{ "nbformat": 4, "nbformat_minor": 5, "metadata": { "kernelspec": { "display_name": "Python 3", "language": "python", "name": "python3" }, "language_info": { "name": "python", "version": "3.10.0" }, "colab": { "provenance": [] } }, "cells": [ { "cell_type": "markdown", "metadata": { "id": "-6t7b7xh2njs" }, "source": "# QML Quantum Data Encoding Tool (Eğitim Amaçlı)\n\n**Veri tipi:** Tabular / Image / Geospatial \n**+ GDPR-KVKK uyumlu Anonymisation / Deanonymisation** \n**Qiskit ile otomatik quantum data encoding**\n\n---\n\n### Özellikler\n1. Başlangıçta veri tipi sorulur (tabular / image / geospatial)\n2. Veri tipine göre otomatik encoding seçimi\n3. Tabular + GDPR/KVKK → hassas özellikler için anonymisation + deanonymisation\n4. Qiskit ile circuit üretimi ve görselleştirme\n\n### Desteklenen Encoding Yöntemleri\n| Encoding | Tabular | Image | Geospatial |\n|----------|---------|-------|------------|\n| Angle Encoding | ✅ | ✅ | ✅ |\n| Amplitude Encoding | ✅ | ✅ | ✅ |\n| ZZFeatureMap | ✅ | - | - |\n| ZFeatureMap | ✅ | - | - |\n\n---\n**Not:** Bu notebook eğitim amaçlıdır. Gerçek KVKK/GDPR uygulamalarında k-anonymity, differential privacy gibi daha güçlü yöntemler kullanılmalıdır.", "id": "-6t7b7xh2njs" }, { "cell_type": "markdown", "metadata": { "id": "ajNwGS4Z2nj1" }, "source": "## 1. Kurulum", "id": "ajNwGS4Z2nj1" }, { "cell_type": "code", "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "AdL4A0xG2nj2", "executionInfo": { "status": "ok", "timestamp": 1787224086352, "user_tz": -180, "elapsed": 26473, "user": { "displayName": "Joseph Volkan ÖZCAN", "userId": "11389530967736397480" } }, "outputId": "934a7ef6-e231-4b75-d483-a4e9545cff71" }, "source": "# Gerekli kütüphaneleri yükle\n!pip install -q qiskit qiskit-aer qiskit-ibm-runtime qiskit-machine-learning \\\n pandas numpy matplotlib seaborn scikit-learn openpyxl pillow\n\nimport numpy as np\nimport pandas as pd\nimport matplotlib.pyplot as plt\nfrom IPython.display import display, Markdown, clear_output\nimport warnings\nwarnings.filterwarnings(\"ignore\")\n\nfrom qiskit import QuantumCircuit\nfrom qiskit.circuit.library import ZZFeatureMap, ZFeatureMap, PauliFeatureMap\nfrom qiskit.quantum_info import Statevector\nfrom qiskit.visualization import plot_bloch_multivector\nfrom qiskit_aer import AerSimulator\n\nprint(\"✅ Kütüphaneler yüklendi.\")", "outputs": [ { "output_type": "stream", "name": "stdout", "text": [ "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m9.8/9.8 MB\u001b[0m \u001b[31m53.1 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m12.4/12.4 MB\u001b[0m \u001b[31m46.8 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2.0/2.0 MB\u001b[0m \u001b[31m19.3 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m282.3/282.3 kB\u001b[0m \u001b[31m6.6 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m412.6/412.6 kB\u001b[0m \u001b[31m12.5 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m120.7/120.7 kB\u001b[0m \u001b[31m5.0 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m94.7/94.7 kB\u001b[0m \u001b[31m3.8 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2.4/2.4 MB\u001b[0m \u001b[31m37.8 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m224.2/224.2 kB\u001b[0m \u001b[31m6.2 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m54.5/54.5 kB\u001b[0m \u001b[31m2.1 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m76.6/76.6 kB\u001b[0m \u001b[31m2.6 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m130.2/130.2 kB\u001b[0m \u001b[31m4.9 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m541.5/541.5 kB\u001b[0m \u001b[31m16.1 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n", "\u001b[?25h✅ Kütüphaneler yüklendi.\n" ] } ], "execution_count": 1, "id": "AdL4A0xG2nj2" }, { "cell_type": "markdown", "metadata": { "id": "A_YaXWU82nj5" }, "source": "## 2. GDPR / KVKK Anonymisation – Deanonymisation Modülü\n\nEğitim amaçlı, **geri dönüşümlü (reversible)** anonymisation.\n\n- `hash_salt`: SHA-256 tabanlı hash (önerilen)\n- `token`: Sıralı token (TOK_0001, TOK_0002, ...)\n- `noise`: Sayısal verilere gürültü ekleme\n- `generalization`: Aralık genelleme", "id": "A_YaXWU82nj5" }, { "cell_type": "code", "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "gTP75Jsb2nj6", "executionInfo": { "status": "ok", "timestamp": 1787224096931, "user_tz": -180, "elapsed": 60, "user": { "displayName": "Joseph Volkan ÖZCAN", "userId": "11389530967736397480" } }, "outputId": "8b84e8d5-6fd7-4efb-8b7e-2348eef04954" }, "source": "class GDPRAnonymizer:\n \"\"\"\n Eğitim amaçlı, geri dönüşümlü (reversible) anonymisation.\n Gerçek üretimde k-anonymity, differential privacy vb. daha güçlü yöntemler kullanılır.\n Bu sınıf, quantum encoding öncesi hassas veriyi korumak ve\n encoding sonrası orijinal değere dönebilmek için tasarlanmıştır.\n \"\"\"\n\n def __init__(self, method=\"hash_salt\"):\n \"\"\"\n method: 'hash_salt' | 'noise' | 'generalization' | 'token'\n \"\"\"\n self.method = method\n self.mapping_ = {} # orijinal → anonymized\n self.reverse_mapping_ = {} # anonymized → orijinal\n self.sensitive_cols_ = []\n self.salt_ = \"QML_EDU_2026_SALT\" # eğitim için sabit; gerçekte rastgele + güvenli saklanır\n\n def fit_transform(self, df: pd.DataFrame, sensitive_cols: list):\n \"\"\"Hassas sütunları anonymize et ve mapping'i sakla.\"\"\"\n self.sensitive_cols_ = sensitive_cols\n df_anon = df.copy()\n\n for col in sensitive_cols:\n if col not in df.columns:\n print(f\"⚠️ Uyarı: '{col}' sütunu bulunamadı, atlanıyor.\")\n continue\n\n original = df[col].astype(str)\n if self.method == \"hash_salt\":\n import hashlib\n anon = original.apply(\n lambda x: hashlib.sha256((x + self.salt_).encode()).hexdigest()[:12]\n )\n elif self.method == \"noise\":\n # Sayısal varsayımı\n if pd.api.types.is_numeric_dtype(df[col]):\n noise = np.random.normal(0, 0.15 * (df[col].std() + 1e-6), size=len(df))\n anon = (df[col] + noise).round(4)\n else:\n print(f\"⚠️ '{col}' sayısal değil, noise uygulanamadı → hash_salt kullanılıyor.\")\n import hashlib\n anon = original.apply(\n lambda x: hashlib.sha256((x + self.salt_).encode()).hexdigest()[:12]\n )\n elif self.method == \"generalization\":\n if pd.api.types.is_numeric_dtype(df[col]):\n bins = pd.qcut(df[col], q=5, duplicates=\"drop\")\n anon = bins.astype(str)\n else:\n anon = original.str[:3] + \"***\"\n elif self.method == \"token\":\n unique_vals = original.unique()\n token_map = {v: f\"TOK_{i:04d}\" for i, v in enumerate(unique_vals)}\n anon = original.map(token_map)\n else:\n raise ValueError(\"Geçersiz method. Seçenekler: hash_salt, noise, generalization, token\")\n\n # Mapping kaydet (deanonymisation için)\n for o, a in zip(original, anon):\n self.mapping_[o] = a\n self.reverse_mapping_[str(a)] = o\n\n df_anon[col] = anon\n\n display(Markdown(\"### 🔒 GDPR/KVKK Anonymisation Uygulandı\"))\n print(f\"Hassas sütunlar: {sensitive_cols}\")\n print(f\"Yöntem: {self.method}\")\n display(df_anon.head())\n return df_anon\n\n def inverse_transform(self, df_anon: pd.DataFrame):\n \"\"\"Anonymized veriyi orijinal değerlere geri çevir (deanonymisation).\"\"\"\n df_orig = df_anon.copy()\n for col in self.sensitive_cols_:\n if col in df_orig.columns:\n df_orig[col] = df_orig[col].astype(str).map(\n lambda x: self.reverse_mapping_.get(x, x)\n )\n display(Markdown(\"### 🔓 Deanonymisation (Geri Dönüşüm) Uygulandı\"))\n display(df_orig.head())\n return df_orig\n\n def show_mapping_sample(self, n=5):\n print(\"Örnek mapping (orijinal → anonymized):\")\n for i, (k, v) in enumerate(list(self.mapping_.items())[:n]):\n print(f\" {k} → {v}\")\n\nprint(\"✅ GDPRAnonymizer sınıfı tanımlandı.\")", "outputs": [ { "output_type": "stream", "name": "stdout", "text": [ "✅ GDPRAnonymizer sınıfı tanımlandı.\n" ] } ], "execution_count": 2, "id": "gTP75Jsb2nj6" }, { "cell_type": "markdown", "metadata": { "id": "h5swr0cv2nj7" }, "source": "## 3. Ana Encoding Motoru (Veri Tipine Göre)\n\nVeri tipine (tabular / image / geospatial) göre otomatik encoding seçer ve Qiskit circuit üretir.", "id": "h5swr0cv2nj7" }, { "cell_type": "code", "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "E98X73p72nj8", "executionInfo": { "status": "ok", "timestamp": 1787224105733, "user_tz": -180, "elapsed": 49, "user": { "displayName": "Joseph Volkan ÖZCAN", "userId": "11389530967736397480" } }, "outputId": "f0839f3f-e1b4-4e11-aacc-8b4996b0805a" }, "source": "class QuantumDataEncoder:\n \"\"\"\n Veri tipine (tabular / image / geospatial) göre\n otomatik encoding seçer ve Qiskit circuit üretir.\n \"\"\"\n\n def __init__(self):\n self.data_type = None\n self.X = None\n self.y = None\n self.n_features = None\n self.anonymizer = None\n self.original_df = None\n self.df_anon = None\n\n # ---------- 1. Veri Tipi Seçimi ----------\n def ask_data_type(self):\n display(Markdown(\"\"\"\n### 📥 Veri Tipinizi Seçin\nLütfen verinizin türünü belirtin:\n\n| Seçenek | Açıklama | Önerilen Encoding |\n|---------|----------|-------------------|\n| **1** | Tabular (CSV, Excel, DataFrame) | Angle / ZZFeatureMap / Amplitude |\n| **2** | Image (görüntü, piksel matrisi) | Amplitude / Angle (küçük) |\n| **3** | Geospatial (koordinat + özellik) | Angle + koordinat özel encoding |\n\n\"\"\"))\n choice = input(\"Seçiminiz (1 / 2 / 3): \").strip()\n mapping = {\"1\": \"tabular\", \"2\": \"image\", \"3\": \"geospatial\"}\n self.data_type = mapping.get(choice, \"tabular\")\n print(f\"✅ Seçilen veri tipi: **{self.data_type.upper()}**\")\n return self.data_type\n\n # ---------- 2. Veri Yükleme & Ön İşleme ----------\n def load_data(self, file_path=None, df=None, image_array=None,\n sensitive_cols=None, anonymize_method=\"hash_salt\",\n target_col=None):\n \"\"\"\n file_path : CSV / Excel yolu\n df : hazır DataFrame\n image_array : numpy array (H, W) veya (H, W, C)\n sensitive_cols : GDPR/KVKK hassas sütun listesi (sadece tabular)\n anonymize_method: 'hash_salt' | 'token' | 'noise' | 'generalization'\n target_col : hedef sütun (opsiyonel)\n \"\"\"\n if self.data_type is None:\n self.ask_data_type()\n\n if self.data_type == \"tabular\":\n if file_path:\n if str(file_path).endswith(\".csv\"):\n df = pd.read_csv(file_path)\n else:\n df = pd.read_excel(file_path)\n if df is None:\n raise ValueError(\"Tabular veri için df veya file_path gerekli.\")\n\n self.original_df = df.copy()\n display(Markdown(\"### 📊 Yüklenen Tabular Veri\"))\n display(df.head())\n\n # GDPR / KVKK\n if sensitive_cols:\n self.anonymizer = GDPRAnonymizer(method=anonymize_method)\n df = self.anonymizer.fit_transform(df, sensitive_cols)\n self.df_anon = df.copy()\n\n # Hedef sütun varsa ayır\n if target_col and target_col in df.columns:\n self.y = df[target_col].values\n df_features = df.drop(columns=[target_col])\n else:\n self.y = None\n df_features = df\n\n # Sadece sayısal özellikler (encoding için)\n numeric_df = df_features.select_dtypes(include=[np.number])\n if numeric_df.empty:\n raise ValueError(\"Sayısal özellik bulunamadı. Encoding için sayısal sütunlar gerekli.\")\n self.X = numeric_df.values.astype(float)\n self.n_features = self.X.shape[1]\n print(f\"✅ Tabular veri hazır: {self.X.shape[0]} örnek, {self.n_features} özellik\")\n\n elif self.data_type == \"image\":\n if image_array is None:\n raise ValueError(\"Image için image_array (numpy) gerekli.\")\n img = np.asarray(image_array, dtype=float)\n # Gri tonlamaya indir (eğitim için)\n if img.ndim == 3:\n img = img.mean(axis=2)\n # Normalize [0, π]\n img = (img - img.min()) / (img.max() - img.min() + 1e-8) * np.pi\n self.X = img.flatten().reshape(1, -1) # tek örnek olarak\n self.n_features = self.X.shape[1]\n print(f\"✅ Image verisi hazır: şekil → {self.n_features} piksel\")\n\n elif self.data_type == \"geospatial\":\n if df is None and file_path is None:\n raise ValueError(\"Geospatial için df veya file_path gerekli.\")\n if file_path:\n df = pd.read_csv(file_path)\n self.original_df = df.copy()\n display(Markdown(\"### 🗺️ Yüklenen Geospatial Veri\"))\n display(df.head())\n numeric_df = df.select_dtypes(include=[np.number])\n self.X = numeric_df.values.astype(float)\n self.n_features = self.X.shape[1]\n print(f\"✅ Geospatial veri hazır: {self.X.shape}\")\n\n # Scaling (Angle / FeatureMap için [0, π] aralığı)\n from sklearn.preprocessing import MinMaxScaler\n self.X = MinMaxScaler(feature_range=(0, np.pi)).fit_transform(self.X)\n return self.X\n\n # ---------- 3. Otomatik Encoding Önerisi ----------\n def recommend_encoding(self, algorithm=\"VQC\", device=\"simulator\",\n n_qubits_available=20):\n \"\"\"\n Veri tipi + algoritma + cihaz kısıtına göre encoding önerir.\n \"\"\"\n n = self.n_features\n qubits_amp = int(np.ceil(np.log2(max(n, 1))))\n\n if self.data_type == \"tabular\":\n if algorithm in [\"QSVM\", \"VQC\"] and n <= n_qubits_available:\n rec = (\"zz_feature_map\",\n \"Tabular + VQC/QSVM → ZZFeatureMap (entanglement + kernel için ideal)\")\n elif n > 16 and qubits_amp <= n_qubits_available:\n rec = (\"amplitude\",\n \"Yüksek boyutlu tabular → Amplitude Encoding (qubit verimli)\")\n else:\n rec = (\"angle\",\n \"Güvenli ve shallow seçenek → Angle Encoding\")\n\n elif self.data_type == \"image\":\n if n <= 64 and device == \"simulator\": # 8x8\n rec = (\"amplitude\",\n \"Küçük görüntü → Amplitude Encoding (piksel → genlik)\")\n elif n <= n_qubits_available:\n rec = (\"angle\",\n \"Görüntü pikselleri → Angle Encoding (her piksel bir rotation)\")\n else:\n rec = (\"amplitude\",\n \"Büyük görüntü → Amplitude (qubit sayısı logaritmik)\")\n\n elif self.data_type == \"geospatial\":\n rec = (\"angle\",\n \"Geospatial (lat/lon + özellik) → Angle Encoding \"\n \"(koordinatlar doğrudan rotation açılarına map edilir)\")\n else:\n rec = (\"angle\", \"Varsayılan → Angle Encoding\")\n\n display(Markdown(f\"### 🎯 Önerilen Encoding: **{rec[0].upper()}**\"))\n print(rec[1])\n return rec[0]\n\n # ---------- 4. Encoding Uygulamaları ----------\n def encode_amplitude(self, sample_idx=0):\n x = self.X[sample_idx].copy()\n norm = np.linalg.norm(x)\n if norm == 0:\n x = np.ones_like(x) / np.sqrt(len(x))\n else:\n x = x / norm\n n_qubits = int(np.ceil(np.log2(len(x))))\n pad = 2**n_qubits - len(x)\n if pad > 0:\n x = np.pad(x, (0, pad))\n qc = QuantumCircuit(n_qubits, name=\"AmplitudeEncoding\")\n qc.initialize(x, range(n_qubits))\n return qc\n\n def encode_angle(self, sample_idx=0, rotation=\"ry\"):\n x = self.X[sample_idx]\n n = len(x)\n # Çok fazla qubit olmasın (eğitim için limit)\n max_q = min(n, 12)\n qc = QuantumCircuit(max_q, name=\"AngleEncoding\")\n for i, val in enumerate(x[:max_q]):\n if rotation == \"ry\":\n qc.ry(val, i)\n elif rotation == \"rx\":\n qc.rx(val, i)\n else:\n qc.rz(val, i)\n return qc\n\n def encode_zz_feature_map(self, sample_idx=0, reps=2, entanglement=\"linear\"):\n n = min(self.n_features, 8) # eğitim için makul limit\n fm = ZZFeatureMap(feature_dimension=n, reps=reps, entanglement=entanglement)\n params = self.X[sample_idx][:n]\n if len(params) == fm.num_parameters:\n qc = fm.assign_parameters(params)\n else:\n # Parametre sayısı uyuşmazsa sadece şablonu döndür\n qc = fm\n return qc\n\n def encode(self, encoding=\"auto\", algorithm=\"VQC\", device=\"simulator\",\n sample_idx=0, **kwargs):\n if encoding == \"auto\":\n encoding = self.recommend_encoding(algorithm, device)\n\n display(Markdown(f\"### 🔧 Uygulanan Encoding: **{encoding.upper()}**\"))\n\n if encoding == \"amplitude\":\n qc = self.encode_amplitude(sample_idx)\n elif encoding == \"angle\":\n qc = self.encode_angle(sample_idx, rotation=kwargs.get(\"rotation\", \"ry\"))\n elif encoding in [\"zz_feature_map\", \"zz\"]:\n qc = self.encode_zz_feature_map(sample_idx, **kwargs)\n else:\n raise ValueError(f\"Desteklenmeyen encoding: {encoding}\")\n\n self._visualize(qc, encoding)\n return qc\n\n def _visualize(self, qc, encoding):\n print(f\"Qubit: {qc.num_qubits} | Depth: {qc.depth()}\")\n try:\n display(qc.draw(\"mpl\", style=\"iqp\", fold=40))\n except Exception as e:\n print(f\"Circuit çizilemedi (text):\\n{qc.draw()}\")\n\n if qc.num_qubits <= 5 and qc.num_parameters == 0:\n try:\n sv = Statevector.from_instruction(qc)\n display(Markdown(\"#### Statevector (ilk 8 genlik)\"))\n print(np.round(sv.data[:8], 4))\n except Exception:\n pass\n\n display(Markdown(f\"\"\"\n**Öğrenci Notu**\n- Veri tipi: `{self.data_type}`\n- Encoding: `{encoding}`\n- GDPR/KVKK anonymisation uygulandı mı? → `{self.anonymizer is not None}`\n\"\"\"))\n\nprint(\"✅ QuantumDataEncoder sınıfı tanımlandı.\")", "outputs": [ { "output_type": "stream", "name": "stdout", "text": [ "✅ QuantumDataEncoder sınıfı tanımlandı.\n" ] } ], "execution_count": 3, "id": "E98X73p72nj8" }, { "cell_type": "markdown", "metadata": { "id": "K3U7SEmE2nj9" }, "source": "## 4. Senaryo A — Tabular + GDPR/KVKK\n\nKişisel veri içeren örnek tabular veri seti ile anonymisation + quantum encoding.", "id": "K3U7SEmE2nj9" }, { "cell_type": "code", "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 1000 }, "id": "QV9CnkF62nj-", "executionInfo": { "status": "ok", "timestamp": 1787224118693, "user_tz": -180, "elapsed": 1237, "user": { "displayName": "Joseph Volkan ÖZCAN", "userId": "11389530967736397480" } }, "outputId": "8a273477-d00d-4b7f-e777-f83958c115e7" }, "source": "print(\"=\" * 60)\nprint(\"SENARYO A: Tabular veri + GDPR/KVKK hassas özellikler\")\nprint(\"=\" * 60)\n\n# Örnek tabular veri (kişisel veri içeren)\nnp.random.seed(42)\ndf_demo = pd.DataFrame({\n \"yas\": np.random.randint(18, 70, 20),\n \"gelir\": np.random.randint(20000, 120000, 20),\n \"sehir_kodu\": np.random.randint(1, 81, 20),\n \"tc_kimlik_ornek\": [f\"10000000{i:02d}\" for i in range(20)], # hassas\n \"email_ornek\": [f\"kisi{i}@ornek.com\" for i in range(20)], # hassas\n \"label\": np.random.randint(0, 2, 20)\n})\n\nencoder = QuantumDataEncoder()\nencoder.data_type = \"tabular\" # soruyu atlamak için doğrudan set\n\n# Hassas sütunları belirt\nsensitive = [\"tc_kimlik_ornek\", \"email_ornek\"]\n\nX = encoder.load_data(\n df=df_demo,\n sensitive_cols=sensitive,\n anonymize_method=\"hash_salt\", # veya \"token\", \"noise\", \"generalization\"\n target_col=\"label\"\n)\n\n# Mapping örneği\nif encoder.anonymizer:\n encoder.anonymizer.show_mapping_sample(n=5)\n\n# Otomatik encoding\nqc = encoder.encode(\n encoding=\"auto\",\n algorithm=\"VQC\",\n device=\"simulator\",\n sample_idx=0,\n reps=2\n)", "outputs": [ { "output_type": "stream", "name": "stdout", "text": [ "============================================================\n", "SENARYO A: Tabular veri + GDPR/KVKK hassas özellikler\n", "============================================================\n" ] }, { "output_type": "display_data", "data": { "text/plain": [ "" ], "text/markdown": "### 📊 Yüklenen Tabular Veri" }, "metadata": {} }, { "output_type": "display_data", "data": { "text/plain": [ " yas gelir sehir_kodu tc_kimlik_ornek email_ornek label\n", "0 56 82955 64 1000000000 kisi0@ornek.com 1\n", "1 69 84925 3 1000000001 kisi1@ornek.com 0\n", "2 46 87969 51 1000000002 kisi2@ornek.com 1\n", "3 32 25311 7 1000000003 kisi3@ornek.com 1\n", "4 60 103104 21 1000000004 kisi4@ornek.com 1" ], "text/html": [ "\n", "
\n", "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
yasgelirsehir_kodutc_kimlik_ornekemail_orneklabel
05682955641000000000kisi0@ornek.com1
1698492531000000001kisi1@ornek.com0
24687969511000000002kisi2@ornek.com1
3322531171000000003kisi3@ornek.com1
460103104211000000004kisi4@ornek.com1
\n", "
\n", "
\n", "\n", "
\n", " \n", "\n", " \n", "\n", " \n", "
\n", "\n", "\n", "
\n", "
\n" ], "application/vnd.google.colaboratory.intrinsic+json": { "type": "dataframe", "summary": "{\n \"name\": \")\",\n \"rows\": 5,\n \"fields\": [\n {\n \"column\": \"yas\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 14,\n \"min\": 32,\n \"max\": 69,\n \"num_unique_values\": 5,\n \"samples\": [\n 69,\n 60,\n 46\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"gelir\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 29881,\n \"min\": 25311,\n \"max\": 103104,\n \"num_unique_values\": 5,\n \"samples\": [\n 84925,\n 103104,\n 87969\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"sehir_kodu\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 27,\n \"min\": 3,\n \"max\": 64,\n \"num_unique_values\": 5,\n \"samples\": [\n 3,\n 21,\n 51\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"tc_kimlik_ornek\",\n \"properties\": {\n \"dtype\": \"string\",\n \"num_unique_values\": 5,\n \"samples\": [\n \"1000000001\",\n \"1000000004\",\n \"1000000002\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"email_ornek\",\n \"properties\": {\n \"dtype\": \"string\",\n \"num_unique_values\": 5,\n \"samples\": [\n \"kisi1@ornek.com\",\n \"kisi4@ornek.com\",\n \"kisi2@ornek.com\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"label\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0,\n \"min\": 0,\n \"max\": 1,\n \"num_unique_values\": 2,\n \"samples\": [\n 0,\n 1\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n }\n ]\n}" } }, "metadata": {} }, { "output_type": "display_data", "data": { "text/plain": [ "" ], "text/markdown": "### 🔒 GDPR/KVKK Anonymisation Uygulandı" }, "metadata": {} }, { "output_type": "stream", "name": "stdout", "text": [ "Hassas sütunlar: ['tc_kimlik_ornek', 'email_ornek']\n", "Yöntem: hash_salt\n" ] }, { "output_type": "display_data", "data": { "text/plain": [ " yas gelir sehir_kodu tc_kimlik_ornek email_ornek label\n", "0 56 82955 64 527649e99f28 dbb6a8c039ff 1\n", "1 69 84925 3 fc387a80a373 bea1c52dfff0 0\n", "2 46 87969 51 2d522d418315 d74efdfa82f1 1\n", "3 32 25311 7 cea0cbe63d36 48f21758d0c7 1\n", "4 60 103104 21 5b2a6cd97c67 433a9685e7c4 1" ], "text/html": [ "\n", "
\n", "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
yasgelirsehir_kodutc_kimlik_ornekemail_orneklabel
0568295564527649e99f28dbb6a8c039ff1
169849253fc387a80a373bea1c52dfff00
24687969512d522d418315d74efdfa82f11
332253117cea0cbe63d3648f21758d0c71
460103104215b2a6cd97c67433a9685e7c41
\n", "
\n", "
\n", "\n", "
\n", " \n", "\n", " \n", "\n", " \n", "
\n", "\n", "\n", "
\n", "
\n" ], "application/vnd.google.colaboratory.intrinsic+json": { "type": "dataframe", "summary": "{\n \"name\": \")\",\n \"rows\": 5,\n \"fields\": [\n {\n \"column\": \"yas\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 14,\n \"min\": 32,\n \"max\": 69,\n \"num_unique_values\": 5,\n \"samples\": [\n 69,\n 60,\n 46\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"gelir\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 29881,\n \"min\": 25311,\n \"max\": 103104,\n \"num_unique_values\": 5,\n \"samples\": [\n 84925,\n 103104,\n 87969\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"sehir_kodu\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 27,\n \"min\": 3,\n \"max\": 64,\n \"num_unique_values\": 5,\n \"samples\": [\n 3,\n 21,\n 51\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"tc_kimlik_ornek\",\n \"properties\": {\n \"dtype\": \"string\",\n \"num_unique_values\": 5,\n \"samples\": [\n \"fc387a80a373\",\n \"5b2a6cd97c67\",\n \"2d522d418315\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"email_ornek\",\n \"properties\": {\n \"dtype\": \"string\",\n \"num_unique_values\": 5,\n \"samples\": [\n \"bea1c52dfff0\",\n \"433a9685e7c4\",\n \"d74efdfa82f1\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"label\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0,\n \"min\": 0,\n \"max\": 1,\n \"num_unique_values\": 2,\n \"samples\": [\n 0,\n 1\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n }\n ]\n}" } }, "metadata": {} }, { "output_type": "stream", "name": "stdout", "text": [ "✅ Tabular veri hazır: 20 örnek, 3 özellik\n", "Örnek mapping (orijinal → anonymized):\n", " 1000000000 → 527649e99f28\n", " 1000000001 → fc387a80a373\n", " 1000000002 → 2d522d418315\n", " 1000000003 → cea0cbe63d36\n", " 1000000004 → 5b2a6cd97c67\n" ] }, { "output_type": "display_data", "data": { "text/plain": [ "" ], "text/markdown": "### 🎯 Önerilen Encoding: **ZZ_FEATURE_MAP**" }, "metadata": {} }, { "output_type": "stream", "name": "stdout", "text": [ "Tabular + VQC/QSVM → ZZFeatureMap (entanglement + kernel için ideal)\n" ] }, { "output_type": "display_data", "data": { "text/plain": [ "" ], "text/markdown": "### 🔧 Uygulanan Encoding: **ZZ_FEATURE_MAP**" }, "metadata": {} }, { "output_type": "stream", "name": "stdout", "text": [ "Qubit: 3 | Depth: 1\n", "Circuit çizilemedi (text):\n", " ┌─────────────────────────────────────┐\n", "q_0: ┤0 ├\n", " │ │\n", "q_1: ┤1 ZZFeatureMap(2.3248,2.0954,2.7434) ├\n", " │ │\n", "q_2: ┤2 ├\n", " └─────────────────────────────────────┘\n" ] }, { "output_type": "display_data", "data": { "text/plain": [ "" ], "text/markdown": "#### Statevector (ilk 8 genlik)" }, "metadata": {} }, { "output_type": "stream", "name": "stdout", "text": [ "[ 0.5142+0.0588j 0.3181-0.1484j 0.4364+0.2077j 0.3276-0.0554j\n", " -0.0324+0.006j 0.1249+0.3396j -0.011 -0.0404j 0.3484-0.0988j]\n" ] }, { "output_type": "display_data", "data": { "text/plain": [ "" ], "text/markdown": "\n**Öğrenci Notu** \n- Veri tipi: `tabular` \n- Encoding: `zz_feature_map` \n- GDPR/KVKK anonymisation uygulandı mı? → `True`\n" }, "metadata": {} } ], "execution_count": 4, "id": "QV9CnkF62nj-" }, { "cell_type": "code", "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 562 }, "id": "66AY5TYI2nj_", "executionInfo": { "status": "ok", "timestamp": 1787224137763, "user_tz": -180, "elapsed": 167, "user": { "displayName": "Joseph Volkan ÖZCAN", "userId": "11389530967736397480" } }, "outputId": "2cc2a982-3348-4d5b-8c87-ee30a9a16559" }, "source": "# Deanonymisation örneği (orijinale geri dönüş)\nif encoder.anonymizer and encoder.df_anon is not None:\n print(\"\\n--- Deanonymisation (orijinale dönüş) ---\")\n df_recovered = encoder.anonymizer.inverse_transform(encoder.df_anon)\n print(\"\\nOrijinal ile karşılaştırma (ilk 3 satır):\")\n display(encoder.original_df[[\"tc_kimlik_ornek\", \"email_ornek\"]].head(3))\n display(df_recovered[[\"tc_kimlik_ornek\", \"email_ornek\"]].head(3))", "outputs": [ { "output_type": "stream", "name": "stdout", "text": [ "\n", "--- Deanonymisation (orijinale dönüş) ---\n" ] }, { "output_type": "display_data", "data": { "text/plain": [ "" ], "text/markdown": "### 🔓 Deanonymisation (Geri Dönüşüm) Uygulandı" }, "metadata": {} }, { "output_type": "display_data", "data": { "text/plain": [ " yas gelir sehir_kodu tc_kimlik_ornek email_ornek label\n", "0 56 82955 64 1000000000 kisi0@ornek.com 1\n", "1 69 84925 3 1000000001 kisi1@ornek.com 0\n", "2 46 87969 51 1000000002 kisi2@ornek.com 1\n", "3 32 25311 7 1000000003 kisi3@ornek.com 1\n", "4 60 103104 21 1000000004 kisi4@ornek.com 1" ], "text/html": [ "\n", "
\n", "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
yasgelirsehir_kodutc_kimlik_ornekemail_orneklabel
05682955641000000000kisi0@ornek.com1
1698492531000000001kisi1@ornek.com0
24687969511000000002kisi2@ornek.com1
3322531171000000003kisi3@ornek.com1
460103104211000000004kisi4@ornek.com1
\n", "
\n", "
\n", "\n", "
\n", " \n", "\n", " \n", "\n", " \n", "
\n", "\n", "\n", "
\n", "
\n" ], "application/vnd.google.colaboratory.intrinsic+json": { "type": "dataframe", "summary": "{\n \"name\": \" display(df_recovered[[\\\"tc_kimlik_ornek\\\", \\\"email_ornek\\\"]]\",\n \"rows\": 5,\n \"fields\": [\n {\n \"column\": \"yas\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 14,\n \"min\": 32,\n \"max\": 69,\n \"num_unique_values\": 5,\n \"samples\": [\n 69,\n 60,\n 46\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"gelir\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 29881,\n \"min\": 25311,\n \"max\": 103104,\n \"num_unique_values\": 5,\n \"samples\": [\n 84925,\n 103104,\n 87969\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"sehir_kodu\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 27,\n \"min\": 3,\n \"max\": 64,\n \"num_unique_values\": 5,\n \"samples\": [\n 3,\n 21,\n 51\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"tc_kimlik_ornek\",\n \"properties\": {\n \"dtype\": \"string\",\n \"num_unique_values\": 5,\n \"samples\": [\n \"1000000001\",\n \"1000000004\",\n \"1000000002\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"email_ornek\",\n \"properties\": {\n \"dtype\": \"string\",\n \"num_unique_values\": 5,\n \"samples\": [\n \"kisi1@ornek.com\",\n \"kisi4@ornek.com\",\n \"kisi2@ornek.com\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"label\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0,\n \"min\": 0,\n \"max\": 1,\n \"num_unique_values\": 2,\n \"samples\": [\n 0,\n 1\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n }\n ]\n}" } }, "metadata": {} }, { "output_type": "stream", "name": "stdout", "text": [ "\n", "Orijinal ile karşılaştırma (ilk 3 satır):\n" ] }, { "output_type": "display_data", "data": { "text/plain": [ " tc_kimlik_ornek email_ornek\n", "0 1000000000 kisi0@ornek.com\n", "1 1000000001 kisi1@ornek.com\n", "2 1000000002 kisi2@ornek.com" ], "text/html": [ "\n", "
\n", "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
tc_kimlik_ornekemail_ornek
01000000000kisi0@ornek.com
11000000001kisi1@ornek.com
21000000002kisi2@ornek.com
\n", "
\n", "
\n", "\n", "
\n", " \n", "\n", " \n", "\n", " \n", "
\n", "\n", "\n", "
\n", "
\n" ], "application/vnd.google.colaboratory.intrinsic+json": { "type": "dataframe", "summary": "{\n \"name\": \" display(df_recovered[[\\\"tc_kimlik_ornek\\\", \\\"email_ornek\\\"]]\",\n \"rows\": 3,\n \"fields\": [\n {\n \"column\": \"tc_kimlik_ornek\",\n \"properties\": {\n \"dtype\": \"string\",\n \"num_unique_values\": 3,\n \"samples\": [\n \"1000000000\",\n \"1000000001\",\n \"1000000002\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"email_ornek\",\n \"properties\": {\n \"dtype\": \"string\",\n \"num_unique_values\": 3,\n \"samples\": [\n \"kisi0@ornek.com\",\n \"kisi1@ornek.com\",\n \"kisi2@ornek.com\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n }\n ]\n}" } }, "metadata": {} }, { "output_type": "display_data", "data": { "text/plain": [ " tc_kimlik_ornek email_ornek\n", "0 1000000000 kisi0@ornek.com\n", "1 1000000001 kisi1@ornek.com\n", "2 1000000002 kisi2@ornek.com" ], "text/html": [ "\n", "
\n", "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
tc_kimlik_ornekemail_ornek
01000000000kisi0@ornek.com
11000000001kisi1@ornek.com
21000000002kisi2@ornek.com
\n", "
\n", "
\n", "\n", "
\n", " \n", "\n", " \n", "\n", " \n", "
\n", "\n", "\n", "
\n", "
\n" ], "application/vnd.google.colaboratory.intrinsic+json": { "type": "dataframe", "summary": "{\n \"name\": \" display(df_recovered[[\\\"tc_kimlik_ornek\\\", \\\"email_ornek\\\"]]\",\n \"rows\": 3,\n \"fields\": [\n {\n \"column\": \"tc_kimlik_ornek\",\n \"properties\": {\n \"dtype\": \"string\",\n \"num_unique_values\": 3,\n \"samples\": [\n \"1000000000\",\n \"1000000001\",\n \"1000000002\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"email_ornek\",\n \"properties\": {\n \"dtype\": \"string\",\n \"num_unique_values\": 3,\n \"samples\": [\n \"kisi0@ornek.com\",\n \"kisi1@ornek.com\",\n \"kisi2@ornek.com\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n }\n ]\n}" } }, "metadata": {} } ], "execution_count": 5, "id": "66AY5TYI2nj_" }, { "cell_type": "markdown", "metadata": { "id": "rka5oOv42nj_" }, "source": "## 5. Senaryo B — Image Verisi", "id": "rka5oOv42nj_" }, { "cell_type": "code", "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 1000 }, "id": "ZcKR18l32nkA", "executionInfo": { "status": "ok", "timestamp": 1787224145525, "user_tz": -180, "elapsed": 376, "user": { "displayName": "Joseph Volkan ÖZCAN", "userId": "11389530967736397480" } }, "outputId": "1d9981ff-320b-4885-8cbc-1cd1961436d1" }, "source": "print(\"\\n\" + \"=\" * 60)\nprint(\"SENARYO B: Image (küçük gri tonlama görüntü)\")\nprint(\"=\" * 60)\n\n# 4x4 örnek görüntü\nnp.random.seed(7)\nimg = np.random.rand(4, 4) * 255\nprint(\"Orijinal görüntü (4x4):\")\nplt.figure(figsize=(4, 4))\nplt.imshow(img, cmap=\"gray\")\nplt.colorbar()\nplt.title(\"Örnek Görüntü\")\nplt.show()\n\nencoder_img = QuantumDataEncoder()\nencoder_img.data_type = \"image\"\nX_img = encoder_img.load_data(image_array=img)\n\nqc_img = encoder_img.encode(\n encoding=\"auto\",\n algorithm=\"VQC\",\n device=\"simulator\",\n sample_idx=0\n)", "outputs": [ { "output_type": "stream", "name": "stdout", "text": [ "\n", "============================================================\n", "SENARYO B: Image (küçük gri tonlama görüntü)\n", "============================================================\n", "Orijinal görüntü (4x4):\n" ] }, { "output_type": "display_data", "data": { "text/plain": [ "
" ], "image/png": "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\n" }, "metadata": {} }, { "output_type": "stream", "name": "stdout", "text": [ "✅ Image verisi hazır: şekil → 16 piksel\n" ] }, { "output_type": "display_data", "data": { "text/plain": [ "" ], "text/markdown": "### 🎯 Önerilen Encoding: **AMPLITUDE**" }, "metadata": {} }, { "output_type": "stream", "name": "stdout", "text": [ "Küçük görüntü → Amplitude Encoding (piksel → genlik)\n" ] }, { "output_type": "display_data", "data": { "text/plain": [ "" ], "text/markdown": "### 🔧 Uygulanan Encoding: **AMPLITUDE**" }, "metadata": {} }, { "output_type": "stream", "name": "stdout", "text": [ "Qubit: 4 | Depth: 1\n", "Circuit çizilemedi (text):\n", " »\n", "q_0: »\n", " »\n", "q_1: »\n", " »\n", "q_2: »\n", " »\n", "q_3: »\n", " »\n", "« ┌──────────────────────────────────────────────────────────────────────────────────────────────┐\n", "«q_0: ┤0 ├\n", "« │ │\n", "«q_1: ┤1 ├\n", "« │ Initialize(0.25,0.25,0.25,0.25,0.25,0.25,0.25,0.25,0.25,0.25,0.25,0.25,0.25,0.25,0.25,0.25) │\n", "«q_2: ┤2 ├\n", "« │ │\n", "«q_3: ┤3 ├\n", "« └──────────────────────────────────────────────────────────────────────────────────────────────┘\n" ] }, { "output_type": "display_data", "data": { "text/plain": [ "" ], "text/markdown": "#### Statevector (ilk 8 genlik)" }, "metadata": {} }, { "output_type": "stream", "name": "stdout", "text": [ "[0.25+0.j 0.25+0.j 0.25+0.j 0.25+0.j 0.25+0.j 0.25+0.j 0.25+0.j 0.25+0.j]\n" ] }, { "output_type": "display_data", "data": { "text/plain": [ "" ], "text/markdown": "\n**Öğrenci Notu** \n- Veri tipi: `image` \n- Encoding: `amplitude` \n- GDPR/KVKK anonymisation uygulandı mı? → `False`\n" }, "metadata": {} } ], "execution_count": 6, "id": "ZcKR18l32nkA" }, { "cell_type": "markdown", "metadata": { "id": "vSQiTR7m2nkB" }, "source": "## 6. Senaryo C — Geospatial Verisi", "id": "vSQiTR7m2nkB" }, { "cell_type": "code", "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 774 }, "id": "KA3rcInh2nkB", "executionInfo": { "status": "ok", "timestamp": 1787224151944, "user_tz": -180, "elapsed": 88, "user": { "displayName": "Joseph Volkan ÖZCAN", "userId": "11389530967736397480" } }, "outputId": "bc9b368b-ede0-487d-efe6-086e48680a40" }, "source": "print(\"\\n\" + \"=\" * 60)\nprint(\"SENARYO C: Geospatial (lat, lon + özellik)\")\nprint(\"=\" * 60)\n\nnp.random.seed(42)\ndf_geo = pd.DataFrame({\n \"latitude\": np.random.uniform(36, 42, 15),\n \"longitude\": np.random.uniform(26, 45, 15),\n \"yagis_mm\": np.random.uniform(0, 100, 15),\n \"sicaklik\": np.random.uniform(5, 35, 15)\n})\n\nencoder_geo = QuantumDataEncoder()\nencoder_geo.data_type = \"geospatial\"\nX_geo = encoder_geo.load_data(df=df_geo)\n\nqc_geo = encoder_geo.encode(\n encoding=\"auto\",\n algorithm=\"VQC\",\n device=\"simulator\",\n sample_idx=0\n)", "outputs": [ { "output_type": "stream", "name": "stdout", "text": [ "\n", "============================================================\n", "SENARYO C: Geospatial (lat, lon + özellik)\n", "============================================================\n" ] }, { "output_type": "display_data", "data": { "text/plain": [ "" ], "text/markdown": "### 🗺️ Yüklenen Geospatial Veri" }, "metadata": {} }, { "output_type": "display_data", "data": { "text/plain": [ " latitude longitude yagis_mm sicaklik\n", "0 38.247241 29.484686 60.754485 24.875669\n", "1 41.704286 31.780603 17.052412 14.351332\n", "2 40.391964 35.970372 6.505159 20.602041\n", "3 39.591951 34.206955 94.888554 21.401308\n", "4 36.936112 31.533354 96.563203 10.545634" ], "text/html": [ "\n", "
\n", "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
latitudelongitudeyagis_mmsicaklik
038.24724129.48468660.75448524.875669
141.70428631.78060317.05241214.351332
240.39196435.9703726.50515920.602041
339.59195134.20695594.88855421.401308
436.93611231.53335496.56320310.545634
\n", "
\n", "
\n", "\n", "
\n", " \n", "\n", " \n", "\n", " \n", "
\n", "\n", "\n", "
\n", "
\n" ], "application/vnd.google.colaboratory.intrinsic+json": { "type": "dataframe", "summary": "{\n \"name\": \")\",\n \"rows\": 5,\n \"fields\": [\n {\n \"column\": \"latitude\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 1.852489497711731,\n \"min\": 36.93611184265462,\n \"max\": 41.7042858384595,\n \"num_unique_values\": 5,\n \"samples\": [\n 41.7042858384595,\n 36.93611184265462,\n 40.39196365086843\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"longitude\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 2.522693428538092,\n \"min\": 29.484685687215244,\n \"max\": 35.97037220101252,\n \"num_unique_values\": 5,\n \"samples\": [\n 31.780602616231217,\n 31.533353663762796,\n 35.97037220101252\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"yagis_mm\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 42.259209016028876,\n \"min\": 6.505159298527952,\n \"max\": 96.56320330745594,\n \"num_unique_values\": 5,\n \"samples\": [\n 17.052412368729154,\n 96.56320330745594,\n 6.505159298527952\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"sicaklik\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 5.784977571055062,\n \"min\": 10.545633665765811,\n \"max\": 24.87566853061946,\n \"num_unique_values\": 5,\n \"samples\": [\n 14.351332282682328,\n 10.545633665765811,\n 20.602040635334326\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n }\n ]\n}" } }, "metadata": {} }, { "output_type": "stream", "name": "stdout", "text": [ "✅ Geospatial veri hazır: (15, 4)\n" ] }, { "output_type": "display_data", "data": { "text/plain": [ "" ], "text/markdown": "### 🎯 Önerilen Encoding: **ANGLE**" }, "metadata": {} }, { "output_type": "stream", "name": "stdout", "text": [ "Geospatial (lat/lon + özellik) → Angle Encoding (koordinatlar doğrudan rotation açılarına map edilir)\n" ] }, { "output_type": "display_data", "data": { "text/plain": [ "" ], "text/markdown": "### 🔧 Uygulanan Encoding: **ANGLE**" }, "metadata": {} }, { "output_type": "stream", "name": "stdout", "text": [ "Qubit: 4 | Depth: 1\n", "Circuit çizilemedi (text):\n", " ┌────────────┐\n", "q_0: ─┤ Ry(1.1713) ├\n", " ┌┴────────────┤\n", "q_1: ┤ Ry(0.58243) ├\n", " └┬────────────┤\n", "q_2: ─┤ Ry(1.9336) ├\n", " ├───────────┬┘\n", "q_3: ─┤ Ry(2.098) ├─\n", " └───────────┘ \n" ] }, { "output_type": "display_data", "data": { "text/plain": [ "" ], "text/markdown": "#### Statevector (ilk 8 genlik)" }, "metadata": {} }, { "output_type": "stream", "name": "stdout", "text": [ "[0.226 +0.j 0.1499+0.j 0.0677+0.j 0.0449+0.j 0.3275+0.j 0.2172+0.j\n", " 0.0982+0.j 0.0651+0.j]\n" ] }, { "output_type": "display_data", "data": { "text/plain": [ "" ], "text/markdown": "\n**Öğrenci Notu** \n- Veri tipi: `geospatial` \n- Encoding: `angle` \n- GDPR/KVKK anonymisation uygulandı mı? → `False`\n" }, "metadata": {} } ], "execution_count": 7, "id": "KA3rcInh2nkB" }, { "cell_type": "markdown", "metadata": { "id": "XYIy2un82nkC" }, "source": "## 7. İnteraktif Başlangıç (Kendi Verinizle)\n\nBu hücreyi çalıştırarak interaktif olarak veri tipi seçebilir, hassas sütunları belirtebilir ve encoding yapabilirsiniz.\n\n> **Colab'da kendi dosyanızı yüklemek için:** \n> `from google.colab import files` → `uploaded = files.upload()` kullanın.", "id": "XYIy2un82nkC" }, { "cell_type": "code", "metadata": { "id": "ANXH9Bjn2nkD", "executionInfo": { "status": "ok", "timestamp": 1787224158864, "user_tz": -180, "elapsed": 9, "user": { "displayName": "Joseph Volkan ÖZCAN", "userId": "11389530967736397480" } } }, "source": "def interactive_start():\n encoder = QuantumDataEncoder()\n data_type = encoder.ask_data_type()\n\n if data_type == \"tabular\":\n print(\"\\nCSV/Excel yüklemek ister misiniz?\")\n print(\"(Colab'da: from google.colab import files; uploaded = files.upload())\")\n print(\"Şimdilik örnek veriyle devam ediyoruz...\\n\")\n\n sens = input(\"GDPR/KVKK hassas sütun adlarını virgülle yazın (yoksa Enter): \").strip()\n sensitive_cols = [s.strip() for s in sens.split(\",\")] if sens else None\n\n # Örnek veri\n df = pd.DataFrame({\n \"feature1\": np.random.randn(30),\n \"feature2\": np.random.randn(30),\n \"feature3\": np.random.randn(30),\n \"ad_soyad\": [f\"Kisi_{i}\" for i in range(30)] # örnek hassas\n })\n if sensitive_cols is None:\n sensitive_cols = [\"ad_soyad\"]\n\n method = input(\"Anonymisation yöntemi (hash_salt / token / noise / generalization) [hash_salt]: \").strip() or \"hash_salt\"\n X = encoder.load_data(df=df, sensitive_cols=sensitive_cols, anonymize_method=method)\n\n elif data_type == \"image\":\n print(\"Image için numpy array verin (şimdilik rastgele 8x8)\")\n img = np.random.rand(8, 8)\n plt.imshow(img, cmap=\"gray\")\n plt.title(\"Rastgele 8x8 Görüntü\")\n plt.show()\n X = encoder.load_data(image_array=img)\n\n else: # geospatial\n df = pd.DataFrame({\n \"lat\": np.random.uniform(36, 42, 20),\n \"lon\": np.random.uniform(26, 45, 20),\n \"deger\": np.random.randn(20)\n })\n X = encoder.load_data(df=df)\n\n # Encoding parametreleri\n algo = input(\"Algoritma (VQC / QSVM / Genel) [VQC]: \").strip() or \"VQC\"\n device = input(\"Cihaz (simulator / ibm_quantum) [simulator]: \").strip() or \"simulator\"\n\n qc = encoder.encode(encoding=\"auto\", algorithm=algo, device=device, sample_idx=0)\n return encoder, qc\n\n\n# Çalıştırmak için aşağıdaki satırın yorumunu kaldırın:\n# encoder, qc = interactive_start()", "outputs": [], "execution_count": 8, "id": "ANXH9Bjn2nkD" }, { "cell_type": "markdown", "metadata": { "id": "r1-hq_B_2nkE" }, "source": "## 8. Encoding Karşılaştırma Tablosu (Eğitim)", "id": "r1-hq_B_2nkE" }, { "cell_type": "code", "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 244 }, "id": "1xRp6MEd2nkE", "executionInfo": { "status": "ok", "timestamp": 1787224175528, "user_tz": -180, "elapsed": 51, "user": { "displayName": "Joseph Volkan ÖZCAN", "userId": "11389530967736397480" } }, "outputId": "69d0a7f1-d1bb-4b29-fd91-cd629429ec81" }, "source": "def encoding_comparison_table():\n data = {\n \"Encoding\": [\"Basis\", \"Amplitude\", \"Angle\", \"Dense Angle\", \"ZZFeatureMap\"],\n \"Qubit (N özellik)\": [\"N\", \"⌈log₂ N⌉\", \"N\", \"⌈N/2⌉\", \"N\"],\n \"Circuit Depth\": [\"O(1)\", \"O(N) ~ yüksek\", \"O(1)\", \"O(1)\", \"O(reps × N)\"],\n \"NISQ Uygunluğu\": [\"Yüksek\", \"Düşük-Orta\", \"Çok Yüksek\", \"Yüksek\", \"Yüksek\"],\n \"En İyi Kullanım\": [\n \"Binary veri, toy örnekler\",\n \"HHL, yüksek boyut, görüntü\",\n \"VQC, QNN, geospatial\",\n \"Qubit tasarrufu\",\n \"QSVM, kernel yöntemleri\"\n ],\n \"Qiskit Desteği\": [\"Manuel\", \"initialize()\", \"Manuel / FeatureMap\", \"Manuel\", \"ZZFeatureMap\"]\n }\n df_comp = pd.DataFrame(data)\n display(Markdown(\"### 📋 Encoding Yöntemleri Karşılaştırması\"))\n display(df_comp)\n\nencoding_comparison_table()", "outputs": [ { "output_type": "display_data", "data": { "text/plain": [ "" ], "text/markdown": "### 📋 Encoding Yöntemleri Karşılaştırması" }, "metadata": {} }, { "output_type": "display_data", "data": { "text/plain": [ " Encoding Qubit (N özellik) Circuit Depth NISQ Uygunluğu \\\n", "0 Basis N O(1) Yüksek \n", "1 Amplitude ⌈log₂ N⌉ O(N) ~ yüksek Düşük-Orta \n", "2 Angle N O(1) Çok Yüksek \n", "3 Dense Angle ⌈N/2⌉ O(1) Yüksek \n", "4 ZZFeatureMap N O(reps × N) Yüksek \n", "\n", " En İyi Kullanım Qiskit Desteği \n", "0 Binary veri, toy örnekler Manuel \n", "1 HHL, yüksek boyut, görüntü initialize() \n", "2 VQC, QNN, geospatial Manuel / FeatureMap \n", "3 Qubit tasarrufu Manuel \n", "4 QSVM, kernel yöntemleri ZZFeatureMap " ], "text/html": [ "\n", "
\n", "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
EncodingQubit (N özellik)Circuit DepthNISQ UygunluğuEn İyi KullanımQiskit Desteği
0BasisNO(1)YüksekBinary veri, toy örneklerManuel
1Amplitude⌈log₂ N⌉O(N) ~ yüksekDüşük-OrtaHHL, yüksek boyut, görüntüinitialize()
2AngleNO(1)Çok YüksekVQC, QNN, geospatialManuel / FeatureMap
3Dense Angle⌈N/2⌉O(1)YüksekQubit tasarrufuManuel
4ZZFeatureMapNO(reps × N)YüksekQSVM, kernel yöntemleriZZFeatureMap
\n", "
\n", "
\n", "\n", "
\n", " \n", "\n", " \n", "\n", " \n", "
\n", "\n", "\n", "
\n", "
\n" ], "application/vnd.google.colaboratory.intrinsic+json": { "type": "dataframe", "summary": "{\n \"name\": \"encoding_comparison_table()\",\n \"rows\": 5,\n \"fields\": [\n {\n \"column\": \"Encoding\",\n \"properties\": {\n \"dtype\": \"string\",\n \"num_unique_values\": 5,\n \"samples\": [\n \"Amplitude\",\n \"ZZFeatureMap\",\n \"Angle\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"Qubit (N \\u00f6zellik)\",\n \"properties\": {\n \"dtype\": \"string\",\n \"num_unique_values\": 3,\n \"samples\": [\n \"N\",\n \"\\u2308log\\u2082 N\\u2309\",\n \"\\u2308N/2\\u2309\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"Circuit Depth\",\n \"properties\": {\n \"dtype\": \"string\",\n \"num_unique_values\": 3,\n \"samples\": [\n \"O(1)\",\n \"O(N) ~ y\\u00fcksek\",\n \"O(reps \\u00d7 N)\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"NISQ Uygunlu\\u011fu\",\n \"properties\": {\n \"dtype\": \"string\",\n \"num_unique_values\": 3,\n \"samples\": [\n \"Y\\u00fcksek\",\n \"D\\u00fc\\u015f\\u00fck-Orta\",\n \"\\u00c7ok Y\\u00fcksek\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"En \\u0130yi Kullan\\u0131m\",\n \"properties\": {\n \"dtype\": \"string\",\n \"num_unique_values\": 5,\n \"samples\": [\n \"HHL, y\\u00fcksek boyut, g\\u00f6r\\u00fcnt\\u00fc\",\n \"QSVM, kernel y\\u00f6ntemleri\",\n \"VQC, QNN, geospatial\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"Qiskit Deste\\u011fi\",\n \"properties\": {\n \"dtype\": \"string\",\n \"num_unique_values\": 4,\n \"samples\": [\n \"initialize()\",\n \"ZZFeatureMap\",\n \"Manuel\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n }\n ]\n}" } }, "metadata": {} } ], "execution_count": 9, "id": "1xRp6MEd2nkE" }, { "cell_type": "markdown", "metadata": { "id": "TxLyHkTd2nkF" }, "source": "## 9. Kendi CSV / Excel Dosyanızı Yükleyin (Colab)\n\nGoogle Colab'da çalışıyorsanız aşağıdaki hücreyi kullanarak kendi dosyanızı yükleyebilirsiniz.", "id": "TxLyHkTd2nkF" }, { "cell_type": "code", "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 1000 }, "id": "bc5UuP8P2nkF", "executionInfo": { "status": "error", "timestamp": 1787224192293, "user_tz": -180, "elapsed": 12469, "user": { "displayName": "Joseph Volkan ÖZCAN", "userId": "11389530967736397480" } }, "outputId": "50bce84e-1fb3-47ac-9178-703a4cd27139" }, "source": "# Colab'da dosya yükleme\ntry:\n from google.colab import files\n print(\"CSV veya Excel dosyanızı seçin (iptal ederseniz örnek veriyle devam edilir):\")\n uploaded = files.upload()\n\n if uploaded:\n filename = list(uploaded.keys())[0]\n print(f\"\\nYüklenen dosya: {filename}\")\n\n encoder_user = QuantumDataEncoder()\n encoder_user.data_type = \"tabular\"\n\n # Hassas sütunları belirtin (örnek)\n # sensitive_user = [\"tc_kimlik\", \"email\"] # kendi sütun adlarınızı yazın\n sensitive_user = None # yoksa None bırakın\n\n X_user = encoder_user.load_data(\n file_path=filename,\n sensitive_cols=sensitive_user,\n anonymize_method=\"hash_salt\"\n )\n\n qc_user = encoder_user.encode(\n encoding=\"auto\",\n algorithm=\"VQC\",\n device=\"simulator\",\n sample_idx=0,\n reps=2\n )\n else:\n print(\"Dosya yüklenmedi → yukarıdaki örnek senaryolarla devam edebilirsiniz.\")\nexcept ImportError:\n print(\"Bu hücre sadece Google Colab'da çalışır.\")\n print(\"Yerel Jupyter'da kendi dosya yolunuzu load_data(file_path=...) ile verin.\")", "outputs": [ { "output_type": "stream", "name": "stdout", "text": [ "CSV veya Excel dosyanızı seçin (iptal ederseniz örnek veriyle devam edilir):\n" ] }, { "output_type": "display_data", "data": { "text/plain": [ "" ], "text/html": [ "\n", " \n", " \n", " Upload widget is only available when the cell has been executed in the\n", " current browser session. Please rerun this cell to enable.\n", " \n", " " ] }, "metadata": {} }, { "output_type": "stream", "name": "stdout", "text": [ "Saving Graniser Grup 17 Ağustos 2026 Dava Raporu.xlsx to Graniser Grup 17 Ağustos 2026 Dava Raporu.xlsx\n", "\n", "Yüklenen dosya: Graniser Grup 17 Ağustos 2026 Dava Raporu.xlsx\n" ] }, { "output_type": "display_data", "data": { "text/plain": [ "" ], "text/markdown": "### 📊 Yüklenen Tabular Veri" }, "metadata": {} }, { "output_type": "display_data", "data": { "text/plain": [ " # Davacı/Alacaklı \\\n", "0 1 VLV Endüstriyel Yangın Tesisat Mühendislik İnş... \n", "1 2 Foçalı Palet \n", "2 3 Foçalı Palet \n", "3 4 Foçalı Palet \n", "4 5 Foçalı Palet \n", "\n", " Davalı/Borçlu \\\n", "0 Graniser Granit Seramik Sanayi ve Ticaret A.Ş. \n", "1 Graniser Granit Seramik Sanayi ve Ticaret A.Ş. \n", "2 Graniser Granit Seramik Sanayi ve Ticaret A.Ş. \n", "3 Graniser Granit Seramik Sanayi ve Ticaret A.Ş. \n", "4 Graniser Granit Seramik Sanayi ve Ticaret A.Ş. \n", "\n", " Mahkeme/İcra Dosya No \\\n", "0 İzmir3. Asliye Ticaret Mahkemesi 2025/776 Esas \n", "1 İzmir 12. İcra Dairesi 2025/8555 Esas \n", "2 İzmir 12. İcra Dairesi 2025/9223 Esas \n", "3 İzmir 7. Asliye ticaret Mahkemesi 2025/1120 Esas \n", "4 İzmir 5. Asliye Ticaret Mahkemesi 2026/329 Esas \n", "\n", " Konu \\\n", "0 İtirazın iptali \n", "1 Örnek 7 İcra Takibi \n", "2 Örnek 7 İcra Takibi \n", "3 İzmir 12. İcra Dairesinin 2025/8555 Esas sayıl... \n", "4 İzmir 12. İcra Dairesinin 2025/9223 Esas sayıl... \n", "\n", " Dava Değeri Tutar PB \\\n", "0 805850,39 (Takip çıkışı tutar) 805850.39 TL \n", "1 5.569.810,37 (Takip çıkışı tutar) 5569810.37 TL \n", "2 ı 1.269.065,81 T (Takip çıkışı tutar) ı 1.269.065,81 T TL \n", "3 5.569.810,37 (Takip çıkışı tutar) 5569810.37 TL \n", "4 1.269.065,81 (Takip çıkışı tutar) 1269065.81 TL \n", "\n", " Son Durum Avukat Görüşü \\\n", "0 Bir sonraki duruşma 13.01.2027 tarihindedir. 0 \n", "1 İcra takibine itiraz edilmiş olup takip durdur... 0 \n", "2 İcra takibine itiraz edilmiş olup takip durdur... 0 \n", "3 Dosya istinaf incelemesindedir. 0 \n", "4 Ön inceleme duruşması 10.11.2026 tarihindedir. 0 \n", "\n", " Ayrılacak Karşılık Notlar \\\n", "0 805850.39 NaN \n", "1 5569810.37 NaN \n", "2 1.269.065,81 T NaN \n", "3 5569810.37 NaN \n", "4 1269065.81 NaN \n", "\n", " Dava Detayı \n", "0 Alacaklı tarafından fatura alacaklarına istina... \n", "1 Alacaklı tarafından cari hesap alacaklarına is... \n", "2 Alacaklı tarafından cari hesap alacaklarına is... \n", "3 Alacaklı tarafından cari hesap alacaklarına is... \n", "4 Alacaklı tarafından cari hesap alacaklarına is... " ], "text/html": [ "\n", "
\n", "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
#Davacı/AlacaklıDavalı/BorçluMahkeme/İcraDosya NoKonuDava DeğeriTutarPBSon DurumAvukat GörüşüAyrılacak KarşılıkNotlarDava Detayı
01VLV Endüstriyel Yangın Tesisat Mühendislik İnş...Graniser Granit Seramik Sanayi ve Ticaret A.Ş.İzmir3. Asliye Ticaret Mahkemesi2025/776 Esasİtirazın iptali805850,39 (Takip çıkışı tutar)805850.39TLBir sonraki duruşma 13.01.2027 tarihindedir.0805850.39NaNAlacaklı tarafından fatura alacaklarına istina...
12Foçalı PaletGraniser Granit Seramik Sanayi ve Ticaret A.Ş.İzmir 12. İcra Dairesi2025/8555 EsasÖrnek 7 İcra Takibi5.569.810,37 (Takip çıkışı tutar)5569810.37TLİcra takibine itiraz edilmiş olup takip durdur...05569810.37NaNAlacaklı tarafından cari hesap alacaklarına is...
23Foçalı PaletGraniser Granit Seramik Sanayi ve Ticaret A.Ş.İzmir 12. İcra Dairesi2025/9223 EsasÖrnek 7 İcra Takibiı 1.269.065,81 T (Takip çıkışı tutar)ı 1.269.065,81 TTLİcra takibine itiraz edilmiş olup takip durdur...01.269.065,81 TNaNAlacaklı tarafından cari hesap alacaklarına is...
34Foçalı PaletGraniser Granit Seramik Sanayi ve Ticaret A.Ş.İzmir 7. Asliye ticaret Mahkemesi2025/1120 Esasİzmir 12. İcra Dairesinin 2025/8555 Esas sayıl...5.569.810,37 (Takip çıkışı tutar)5569810.37TLDosya istinaf incelemesindedir.05569810.37NaNAlacaklı tarafından cari hesap alacaklarına is...
45Foçalı PaletGraniser Granit Seramik Sanayi ve Ticaret A.Ş.İzmir 5. Asliye Ticaret Mahkemesi2026/329 Esasİzmir 12. İcra Dairesinin 2025/9223 Esas sayıl...1.269.065,81 (Takip çıkışı tutar)1269065.81TLÖn inceleme duruşması 10.11.2026 tarihindedir.01269065.81NaNAlacaklı tarafından cari hesap alacaklarına is...
\n", "
\n", "
\n", "\n", "
\n", " \n", "\n", " \n", "\n", " \n", "
\n", "\n", "\n", "
\n", "
\n" ], "application/vnd.google.colaboratory.intrinsic+json": { "type": "dataframe", "summary": "{\n \"name\": \" print(\\\"Yerel Jupyter'da kendi dosya yolunuzu load_data(file_path=\",\n \"rows\": 5,\n \"fields\": [\n {\n \"column\": \"#\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 1,\n \"min\": 1,\n \"max\": 5,\n \"num_unique_values\": 5,\n \"samples\": [\n 2,\n 5,\n 3\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"Davac\\u0131/Alacakl\\u0131\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 2,\n \"samples\": [\n \"Fo\\u00e7al\\u0131 Palet\",\n \"VLV End\\u00fcstriyel Yang\\u0131n Tesisat M\\u00fchendislik \\u0130n\\u015faat Otomasyon \\u0130thalat \\u0130hracat Sanayi ve Ticaret Limited \\u015eirketi \"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"Daval\\u0131/Bor\\u00e7lu\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 1,\n \"samples\": [\n \"Graniser Granit Seramik Sanayi ve Ticaret A.\\u015e.\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"Mahkeme/\\u0130cra\",\n \"properties\": {\n \"dtype\": \"string\",\n \"num_unique_values\": 4,\n \"samples\": [\n \"\\u0130zmir 12. \\u0130cra Dairesi\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"Dosya No\",\n \"properties\": {\n \"dtype\": \"string\",\n \"num_unique_values\": 5,\n \"samples\": [\n \"2025/8555 Esas \"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"Konu\",\n \"properties\": {\n \"dtype\": \"string\",\n \"num_unique_values\": 4,\n \"samples\": [\n \"\\u00d6rnek 7 \\u0130cra Takibi\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"Dava De\\u011feri\",\n \"properties\": {\n \"dtype\": \"string\",\n \"num_unique_values\": 4,\n \"samples\": [\n \"5.569.810,37 (Takip \\u00e7\\u0131k\\u0131\\u015f\\u0131 tutar)\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"Tutar\",\n \"properties\": {\n \"dtype\": \"string\",\n \"num_unique_values\": 4,\n \"samples\": [\n 5569810.37\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"PB\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 1,\n \"samples\": [\n \"TL\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"Son Durum\",\n \"properties\": {\n \"dtype\": \"string\",\n \"num_unique_values\": 4,\n \"samples\": [\n \"\\u0130cra takibine itiraz edilmi\\u015f olup takip durdurulmu\\u015ftur.\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"Avukat G\\u00f6r\\u00fc\\u015f\\u00fc\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0,\n \"min\": 0,\n \"max\": 0,\n \"num_unique_values\": 1,\n \"samples\": [\n 0\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"Ayr\\u0131lacak Kar\\u015f\\u0131l\\u0131k\",\n \"properties\": {\n \"dtype\": \"string\",\n \"num_unique_values\": 4,\n \"samples\": [\n 5569810.37\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"Notlar\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": null,\n \"min\": null,\n \"max\": null,\n \"num_unique_values\": 0,\n \"samples\": [],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"Dava Detay\\u0131\",\n \"properties\": {\n \"dtype\": \"string\",\n \"num_unique_values\": 4,\n \"samples\": [],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n }\n ]\n}" } }, "metadata": {} }, { "output_type": "stream", "name": "stdout", "text": [ "✅ Tabular veri hazır: 6 örnek, 3 özellik\n" ] }, { "output_type": "display_data", "data": { "text/plain": [ "" ], "text/markdown": "### 🎯 Önerilen Encoding: **ZZ_FEATURE_MAP**" }, "metadata": {} }, { "output_type": "stream", "name": "stdout", "text": [ "Tabular + VQC/QSVM → ZZFeatureMap (entanglement + kernel için ideal)\n" ] }, { "output_type": "display_data", "data": { "text/plain": [ "" ], "text/markdown": "### 🔧 Uygulanan Encoding: **ZZ_FEATURE_MAP**" }, "metadata": {} }, { "output_type": "error", "ename": "RuntimeError", "evalue": "Binding to NaN.", "traceback": [ "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", "\u001b[0;31mRuntimeError\u001b[0m Traceback (most recent call last)", "\u001b[0;32m/tmp/ipykernel_1936/1536692266.py\u001b[0m in \u001b[0;36m\u001b[0;34m()\u001b[0m\n\u001b[1;32m 22\u001b[0m )\n\u001b[1;32m 23\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 24\u001b[0;31m qc_user = encoder_user.encode(\n\u001b[0m\u001b[1;32m 25\u001b[0m \u001b[0mencoding\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m\"auto\"\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 26\u001b[0m \u001b[0malgorithm\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m\"VQC\"\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", "\u001b[0;32m/tmp/ipykernel_1936/805529563.py\u001b[0m in \u001b[0;36mencode\u001b[0;34m(self, encoding, algorithm, device, sample_idx, **kwargs)\u001b[0m\n\u001b[1;32m 210\u001b[0m \u001b[0mqc\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mencode_angle\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0msample_idx\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mrotation\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mget\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"rotation\"\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m\"ry\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 211\u001b[0m \u001b[0;32melif\u001b[0m \u001b[0mencoding\u001b[0m \u001b[0;32min\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m\"zz_feature_map\"\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m\"zz\"\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 212\u001b[0;31m \u001b[0mqc\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mencode_zz_feature_map\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0msample_idx\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 213\u001b[0m \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 214\u001b[0m \u001b[0;32mraise\u001b[0m \u001b[0mValueError\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mf\"Desteklenmeyen encoding: {encoding}\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", "\u001b[0;32m/tmp/ipykernel_1936/805529563.py\u001b[0m in \u001b[0;36mencode_zz_feature_map\u001b[0;34m(self, sample_idx, reps, entanglement)\u001b[0m\n\u001b[1;32m 192\u001b[0m \u001b[0mparams\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mX\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0msample_idx\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0mn\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 193\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mparams\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0mfm\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mnum_parameters\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 194\u001b[0;31m \u001b[0mqc\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mfm\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0massign_parameters\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mparams\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 195\u001b[0m \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 196\u001b[0m \u001b[0;31m# Parametre sayısı uyuşmazsa sadece şablonu döndür\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", "\u001b[0;32m/usr/local/lib/python3.12/dist-packages/qiskit/circuit/library/n_local/n_local.py\u001b[0m in \u001b[0;36massign_parameters\u001b[0;34m(self, parameters, inplace, **kwargs)\u001b[0m\n\u001b[1;32m 1073\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_build\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1074\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1075\u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0msuper\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0massign_parameters\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mparameters\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0minplace\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0minplace\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 1076\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1077\u001b[0m def _parameterize_block(\n", "\u001b[0;32m/usr/local/lib/python3.12/dist-packages/qiskit/circuit/quantumcircuit.py\u001b[0m in \u001b[0;36massign_parameters\u001b[0;34m(self, parameters, inplace, flat_input, strict)\u001b[0m\n\u001b[1;32m 5070\u001b[0m \u001b[0mtarget\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_data\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0massign_parameters_mapping\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mraw_mapping\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 5071\u001b[0m \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 5072\u001b[0;31m \u001b[0mtarget\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_data\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0massign_parameters_iterable\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mparameters\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 5073\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 5074\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0;32mNone\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0minplace\u001b[0m \u001b[0;32melse\u001b[0m \u001b[0mtarget\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", "\u001b[0;31mRuntimeError\u001b[0m: Binding to NaN." ] } ], "execution_count": 10, "id": "bc5UuP8P2nkF" }, { "cell_type": "markdown", "metadata": { "id": "FG5cUOz92nkG" }, "source": "---\n\n## Özet Akış (Öğrenciler İçin)\n\n| Adım | Ne olur? |\n|------|----------|\n| 1 | **Veri tipi sorulur** → tabular / image / geospatial |\n| 2 | Veri yüklenir |\n| 3 | **Tabular ise** → hassas sütunlar belirtilir → **anonymisation** uygulanır |\n| 4 | Veri tipine + algoritmaya + cihaza göre **encoding otomatik seçilir** |\n| 5 | Qiskit circuit üretilir ve çizilir |\n| 6 | İstenirse **deanonymisation** ile orijinal veri geri getirilir |\n\n### GDPR/KVKK Notu (Eğitim)\n\n- Bu notebook’taki anonymisation **geri dönüşümlüdür** (mapping saklanır) → eğitim ve test için uygundur.\n- Gerçek KVKK/GDPR uyumlu sistemlerde:\n - k-anonymity, l-diversity, differential privacy,\n - irreversible hashing / tokenization,\n - veri minimizasyonu \n tercih edilir.\n- Quantum encoding **anonymize edilmiş** veri üzerinde yapılır; böylece hassas bilgi Hilbert uzayına doğrudan girmez.\n\n---\n\n**Hazırlayan:** QML Eğitim Araçları \n**Tarih:** 2026 \n**Lisans:** Eğitim amaçlı serbest kullanım", "id": "FG5cUOz92nkG" } ] }