{ "cells": [ { "cell_type": "markdown", "id": "ce50dd47", "metadata": {}, "source": [ "**Author:** Dr. Mallarapu \n", "**Created:** 2026-07-27 \n", "**Course:** SEAS 8414 \u2014 Security Analytics\n", "\n", "---\n", "\n", "### Goal of this notebook\n", "\n", "Train and audit anomaly detectors on HDFS log sessions represented as event-template counts.\n", "\n", "### What you will learn\n", "\n", "1. Read a majority-class baseline before trusting any accuracy figure.\n", "2. Find the strongest single feature, then test it by dropping it and refitting.\n", "3. Tell duplicate inflation apart from genuine signal.\n", "4. Report per-group recall, because the rare classes carry the risk.\n", "5. See what a count representation discards, namely event order.\n", "\n", "### Where this connects to the course text\n", "\n", "The text builds a defence pipeline; this notebook trains a classifier and audits it. The links below are to specific chapter objectives that share an *analytic move*, not to matching subject matter.\n", "\n", "- **Chapter 10: Active Deception & Threat Hunting** \u2014 Learning objective 6 (section 10.1) places a claim on the **attribution ladder** and corrects for **dependence among rule hits**. The same rule stops us equating one feature with the label.\n", "- **Chapter 11: Formal Protocol Verification** \u2014 Section **11.1.2**, titled *Proved, tested, and hoped*, asks you to separate exactly those three. (Chapter 11 lists its objectives in \u00a711.0, not \u00a711.1 as the other chapters do.) The ablation does that job here: it tests whether the headline survives.\n", "- **Chapter 12: Autonomous Remediation and Safety Verification** \u2014 Learning objective 1 (section 12.1) assembles evidence into a **safety case** with stated assumptions. Section 13 is that safety case for a model score.\n", "\n", "---" ] }, { "cell_type": "markdown", "id": "b214ead5", "metadata": {}, "source": [ "# System-Log Anomaly Detection on HDFS (Loghub)\n", "### Model comparison + per-anomaly-type recall + validity audit (575k block sessions)\n", "\n", "**Abstract:** The HDFS log dataset (Xu et al., 2009) is the classic benchmark for **console-log anomaly detection**. 11M raw log lines are parsed into event templates and grouped by block id into 575,061 sessions. Each session is a **count vector over 29 event templates** (28 of them vary and are used as features), labelled normal (Success) or anomalous (Fail). We compare four learners on the event-count matrix, report recall per anomaly type, and audit the result. A **new domain** \u2014 log analytics \u2014 distinct from the network/host/fraud datasets elsewhere in this series." ] }, { "cell_type": "markdown", "id": "65314daa", "metadata": {}, "source": [ "## 1. Research problem\n", "\n", "**Task:** Classify an HDFS block session as normal or anomalous from how many times each of 29 log-event templates fired. This is the workhorse of AIOps / reliability engineering. Operators cannot read 11M log lines, so the model must flag the ~3% of sessions worth investigating without drowning them in false positives." ] }, { "cell_type": "markdown", "id": "8833c948", "metadata": {}, "source": [ "## 2. Literature review\n", "\n", "- **Xu, Huang, Fox, Patterson & Jordan (2009)** \u2014 *Detecting Large-Scale System Problems by Mining Console Logs* (SOSP): the HDFS dataset and PCA-based detection.\n", "- **Zhu, He, He, Liu & Lyu (2023)** \u2014 *Loghub*: the log-dataset collection this HDFS release comes from.\n", "- **He, Zhu, Zheng & Lyu (2017)** \u2014 *Drain*: an online log parser for turning raw lines into templates. A background method here. It did **not** produce the counts used in this notebook \u2014 see section 3.\n", "- **Du, Li, Zheng & Srikumar (2017)** \u2014 *DeepLog* (ACM CCS): LSTM log anomaly detection.\n", "- **Sommer & Paxson (2010)** \u2014 the closed-world ML critique.\n", "\n", "**Related approaches and their known caveats** \u2014 drawn from the wider literature; these are **not** measurements reproduced on this exact corpus:\n", "\n", "| Reported approach | Known caveat |\n", "|---|---|\n", "| Xu et al. (2009) \u2014 PCA on event-count vectors | unsupervised; the count matrix makes normal/anomaly nearly separable |\n", "| DeepLog (Du et al., 2017) \u2014 sequence models | uses event ORDER, not just counts; catches order-only anomalies this matrix cannot |\n" ] }, { "cell_type": "markdown", "id": "8b57c014", "metadata": {}, "source": [ "## 3. Dataset provenance & honesty caveats\n", "\n", "| Property | Value |\n", "|---|---|\n", "| Source | Kaggle `tamaniwilliams/hdfs-v1-loghub-dataset-archive` (Loghub HDFS_v1) |\n", "| Rows | 575,061 block sessions (Event_occurrence_matrix) |\n", "| Label | Success (normal) vs Fail (anomaly, ~2.9%); family = anomaly Type code |\n", "| Access | Kaggle API token required |\n", "\n", "**Who parsed these logs: not this notebook, and not Drain.** The archive ships the count matrix already built. Section 6 downloads `Event_occurrence_matrix.csv` and reads it straight into a dataframe. No log parser runs anywhere in this pipeline. The same archive ships `HDFS.log_templates.csv`, listing the templates the columns count. Three facts, deliberately kept apart:\n", "\n", "1. Loghub (Zhu et al., 2023) states it preprocessed this release for research use. Its README does not name the parser that cut the shipped templates.\n", "2. The underlying log set and its block labels come from Xu et al. (2009). Their published *method* recovered log structure from Hadoop source code, not from a data-driven online parser. That is a fact about their method, not a claim that these exact template strings are theirs.\n", "3. Drain (He et al., 2017) is the tool you would reach for to rebuild such a matrix from raw lines yourself. It is a method reference here, never a credit for these numbers.\n", "\n", "One-line check once the download finishes: `!ls /tmp/kg_hdfs`, then open `HDFS.log_templates.csv`.\n", "\n", "**Honestly:** the features are event-template **counts**. This task therefore captures volume/frequency anomalies, but is blind to pure **ordering** anomalies (which DeepLog targets). The anomaly `Type` is label-derived and dropped from features. Anomalies are ~3%, so accuracy is inflated \u2014 per-type recall and the false-positive rate are the honest metrics.\n", "\n", "### Before you run this: getting the data\n", "\n", "This notebook downloads its own data on the first run, then caches it. You do not fetch anything by hand.\n", "\n", "**Dataset:** Kaggle `tamaniwilliams/hdfs-v1-loghub-dataset-archive` -> `/tmp/kg_hdfs`. It is about **2 GB** on disk.\n", "\n", "**One-time setup.** Sign in at kaggle.com, open **Settings**, and under **API** choose **Create New Token**. Kaggle hands you a `kaggle.json` file. This notebook does *not* read that file. It reads a plain key file, so convert it once:\n", "\n", "```bash\n", "mkdir -p ~/.kaggle\n", "python3 -c \"import json;print(json.load(open('kaggle.json'))['key'],end='')\" > ~/.kaggle/access_token\n", "chmod 600 ~/.kaggle/access_token\n", "```\n", "\n", "Never paste the token into a cell, a commit, or a screenshot. If it leaks, revoke it from the same Settings page.\n", "\n", "**If the loader fails:**\n", "\n", "- `FileNotFoundError: ~/.kaggle/access_token` - you created `kaggle.json` but not the key file. Run the command above.\n", "- `401 Unauthorized` - the key is wrong, or a trailing newline crept in.\n", "- `403 Forbidden` - open the dataset page on Kaggle while signed in, accept its terms, then re-run the cell.\n", "\n", "The cache sits under `/tmp`, which macOS clears on reboot. To keep it, move the folder somewhere durable and symlink it back. Do **not** edit the path in the code cell below: that changes a code cell and invalidates the stored outputs you are reviewing." ] }, { "cell_type": "markdown", "id": "6b7bfd17", "metadata": {}, "source": [ "## 4. Solution design\n", "\n", "The methodology is deliberately two-track. We *earn* a headline score with standard modelling, then *interrogate* it with a validity audit. Only a verdict that survives both is reported. The diagram below is the shape of every notebook in this series." ] }, { "cell_type": "markdown", "id": "25ecc50c", "metadata": {}, "source": [ "**Figure 4.1 \u2014 Solution design (methodology).**\n", "\n", "\"Figure" ] }, { "cell_type": "markdown", "id": "94ad91e3", "metadata": {}, "source": [ "## 5. Implementation architecture\n", "\n", "Five stages \u2014 ingestion, preprocessing, modelling, evaluation, and a parallel validity-audit path \u2014 feed a single graded results ledger. Leakage defences (dropping label-derived and identifier columns) live in preprocessing, before any model sees the data." ] }, { "cell_type": "markdown", "id": "06f744b9", "metadata": {}, "source": [ "**Figure 5.1 \u2014 Implementation architecture.**\n", "\n", "\"Figure" ] }, { "cell_type": "markdown", "id": "e4dee67e", "metadata": {}, "source": [ "## 6. Data acquisition & preparation\n", "\n", "Every line below is commented so a student can re-run and modify each step. The cell ends by producing the standard analysis variables: `df`, `X` (clean numeric features), `y` (binary label), `feat` (feature names), and `family`. `family` is the per-group label used for the recall breakdown. It is an attack family on the intrusion corpora, but a transaction type, merchant category or malware category on the fraud/malware ones." ] }, { "cell_type": "code", "execution_count": 1, "id": "67591a05", "metadata": {}, "outputs": [], "source": [ "%matplotlib inline\n", "import time, warnings; warnings.filterwarnings('ignore') # keep output clean\n", "import numpy as np, pandas as pd # numerics + dataframes\n", "import matplotlib.pyplot as plt # static plots (embed in HTML+PDF)\n", "plt.rcParams['figure.dpi'] = 120 # crisp figures\n", "RANDOM_STATE = 0 # single seed used everywhere\n", "np.random.seed(RANDOM_STATE) # reproducible sampling\n", "NEG_WORD, POS_WORD = 'benign', 'attack' # class names (overridden by some loaders)" ] }, { "cell_type": "code", "execution_count": 2, "id": "ce6ab77c", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "loaded 575,061 HDFS block sessions x 28 event-count features; anomaly rate 0.0293\n" ] } ], "source": [ "import os, glob\n", "# HDFS log-anomaly benchmark (Xu et al., 2009; distributed via Loghub, Zhu et al. 2023): 575,061 HDFS\n", "# block sessions, each a COUNT VECTOR over 29 log-event templates (E1..E29), labelled Success/Fail\n", "# (anomalous). A NEW domain for the series \u2014 system-log anomaly detection, not network/host/fraud.\n", "os.environ.setdefault('KAGGLE_KEY', open(os.path.expanduser('~/.kaggle/access_token')).read().strip())\n", "DEST = '/tmp/kg_hdfs'; os.makedirs(DEST, exist_ok=True)\n", "if not glob.glob(DEST + '/**/Event_occurrence_matrix.csv', recursive=True):\n", " import kaggle; kaggle.api.authenticate()\n", " print('downloading HDFS loghub (one-time)...')\n", " kaggle.api.dataset_download_files('tamaniwilliams/hdfs-v1-loghub-dataset-archive', path=DEST, unzip=True, quiet=True)\n", "f = [x for x in glob.glob(DEST + '/**/*.csv', recursive=True) if os.path.basename(x) == 'Event_occurrence_matrix.csv'][0]\n", "df = pd.read_csv(f, low_memory=False); df.columns = [str(c).strip() for c in df.columns]\n", "assert len(df) >= 400_000, f'floor not met: {len(df):,}'\n", "df['y'] = (df['Label'].astype(str).str.strip().str.lower() != 'success').astype(int) # Fail = anomaly = 1\n", "# family = anomaly Type code for anomalies, 'normal' for success -> per-anomaly-type recall.\n", "df['family'] = np.where(df['y'] == 1, 'type_' + df['Type'].astype(str), 'normal')\n", "# Drop the block id, the label and the label-derived anomaly Type. Keep ONLY the E* event counts.\n", "DROP = ['BlockId', 'Label', 'Type', 'y', 'family']\n", "feat = [c for c in df.columns if c not in DROP]\n", "from sklearn.preprocessing import LabelEncoder\n", "X = df[feat].copy()\n", "idlike = [c for c in X.select_dtypes(include='object').columns if X[c].nunique() > 0.5*len(X)]\n", "X = X.drop(columns=idlike) # drop id/timestamp-like leaky columns\n", "for c in X.select_dtypes(include='object').columns:\n", " X[c] = LabelEncoder().fit_transform(X[c].astype(str))\n", "X = X.apply(pd.to_numeric, errors='coerce').replace([np.inf,-np.inf],np.nan).fillna(0.0)\n", "X = X.clip(-1e15, 1e15); X = X.loc[:, X.nunique() > 1] # float32-safe; drop constants\n", "import re\n", "_seen, _cols = {}, []\n", "for _c in X.columns: # unique LightGBM-safe names\n", " _c = re.sub(r'[^0-9A-Za-z_]+', '_', str(_c)).strip('_') or 'f'\n", " _seen[_c] = _seen.get(_c, -1) + 1\n", " _cols.append(_c if _seen[_c] == 0 else f'{_c}_{_seen[_c]}')\n", "X.columns = _cols; feat = list(X.columns)\n", "y = df['y'].to_numpy(); family = df['family'].to_numpy()\n", "print(f'loaded {len(df):,} HDFS block sessions x {len(feat)} event-count features; anomaly rate {y.mean():.4f}')" ] }, { "cell_type": "markdown", "id": "65a9bb6e", "metadata": {}, "source": [ "## 7. Exploratory data analysis\n", "\n", "**Read the class words with care:** The setup cell fixes series-wide class names: `NEG_WORD, POS_WORD = 'benign', 'attack'`. The HDFS loader never overrides them. So the plots below are titled *Class balance (benign vs attack)* and *Top attack families*. The confusion-matrix ticks in section 9 carry the same words, and so does the printed false-positive line (\"benign flagged\").\n", "\n", "Those words are wrong for this corpus. Read `benign` as **Success** \u2014 a block session that completed. Read `attack` as **Fail** \u2014 an anomalous block session. Nothing here is an adversary. These are HDFS block failures, and the `family` groups are anomaly `Type` codes, not attack families.\n", "\n", "This is a code-level defect in a shared template, not a finding about the data. The outputs come from a long run and were not regenerated, so the labels stand as printed. One-line check: `print(NEG_WORD, POS_WORD)` after the loader cell." ] }, { "cell_type": "code", "execution_count": 3, "id": "fb209dbc", "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# --- EDA 1: class balance and the attack-family mix ---\n", "fig, ax = plt.subplots(1, 2, figsize=(11, 4))\n", "df['y'].map({0:NEG_WORD,1:POS_WORD}).value_counts().plot.bar( # counts per class\n", " ax=ax[0], color=['#2a9d8f','#e76f51']); ax[0].set_yscale('log')\n", "ax[0].set_title(f'Class balance ({NEG_WORD} vs {POS_WORD})'); ax[0].set_ylabel('records (log)')\n", "df.loc[df.y==1,'family'].value_counts().head(8).plot.barh( # top attack families\n", " ax=ax[1], color='#e76f51'); ax[1].invert_yaxis(); ax[1].set_title('Top attack families')\n", "plt.tight_layout(); plt.show()" ] }, { "cell_type": "code", "execution_count": 4, "id": "311df064", "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# --- EDA 2: feature correlation + a 2-D PCA projection ---\n", "from sklearn.preprocessing import StandardScaler # scale before PCA\n", "from sklearn.decomposition import PCA\n", "fig, ax = plt.subplots(1, 2, figsize=(12, 5))\n", "topv = X[feat].var().sort_values().tail(12).index # 12 highest-variance features\n", "im = ax[0].imshow(X[topv].corr(), cmap='coolwarm', vmin=-1, vmax=1) # correlation heatmap\n", "ax[0].set_xticks(range(len(topv))); ax[0].set_xticklabels(topv, rotation=90, fontsize=7)\n", "ax[0].set_yticks(range(len(topv))); ax[0].set_yticklabels(topv, fontsize=7)\n", "ax[0].set_title('Feature correlation (top-variance)'); fig.colorbar(im, ax=ax[0], shrink=0.7)\n", "samp = X.sample(min(5000, len(X)), random_state=RANDOM_STATE) # subsample for a fast PCA\n", "pc = PCA(n_components=2).fit_transform(StandardScaler().fit_transform(samp))\n", "ys = y[samp.index] # aligned labels for coloring\n", "for lab,c in [(0,'#2a9d8f'),(1,'#e76f51')]:\n", " ax[1].scatter(pc[ys==lab,0], pc[ys==lab,1], s=4, alpha=0.4, color=c,\n", " label={0:NEG_WORD,1:POS_WORD}[lab])\n", "ax[1].set_title('PCA projection (2 components)'); ax[1].legend(); ax[1].set_xlabel('PC1'); ax[1].set_ylabel('PC2')\n", "plt.tight_layout(); plt.show()" ] }, { "cell_type": "markdown", "id": "c6a1d19c", "metadata": {}, "source": [ "## 8. Model comparison\n", "\n", "Four diverse learners share one held-out split, ranked by ROC-AUC.\n", "\n", "**Two honesty guards print with the table:**\n", "\n", "1. The models train on a *stratified subsample* of at most 120,000 rows. The full row count is printed above. So every score here is a subsample number, not a full-corpus claim.\n", "2. The **majority-class baseline accuracy** appears *inside* the ranking table. On imbalanced data, 0.99 accuracy can be worse than always guessing the majority class. Judge each model against that baseline, not against 0.5." ] }, { "cell_type": "code", "execution_count": 5, "id": "e3cd45e0", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "materialized 575,061 rows | trained on 120,000 (stratified subsample) | held-out 143,766\n", "MAJORITY-CLASS BASELINE accuracy = 0.9707 (any model must beat THIS, not 0.5, to be interesting)\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "best model: RandomForest\n" ] }, { "data": { "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", "
modelaccuracyroc_auctrain_s
0RandomForest0.9998330.9999830.2
1XGBoost0.9997910.9999130.2
2LightGBM0.9998470.9998311.1
3LogisticRegression0.9992560.9986810.1
4MajorityBaseline0.9707000.5000000.0
\n", "
" ], "text/plain": [ " model accuracy roc_auc train_s\n", "0 RandomForest 0.999833 0.999983 0.2\n", "1 XGBoost 0.999791 0.999913 0.2\n", "2 LightGBM 0.999847 0.999831 1.1\n", "3 LogisticRegression 0.999256 0.998681 0.1\n", "4 MajorityBaseline 0.970700 0.500000 0.0" ] }, "execution_count": 5, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# --- Model comparison: four learners on the same held-out split ---\n", "from sklearn.model_selection import train_test_split\n", "from sklearn.linear_model import LogisticRegression\n", "from sklearn.ensemble import RandomForestClassifier\n", "from sklearn.metrics import accuracy_score, roc_auc_score\n", "import xgboost as xgb, lightgbm as lgb\n", "\n", "# Stratified split keeps the class ratio in both halves.\n", "Xtr, Xte, ytr, yte = train_test_split(X, y, test_size=0.25, random_state=RANDOM_STATE, stratify=y)\n", "from sklearn.pipeline import make_pipeline\n", "from sklearn.preprocessing import StandardScaler\n", "N_MATERIALIZED = len(y) # the full corpus we loaded (see printed count)\n", "# HONEST DISCLOSURE: we do NOT train on all N. We fit on a STRATIFIED subsample (<=120k) because\n", "# these learners saturate long before then on this data. Every headline below is a SUBSAMPLE\n", "# number, not a full-corpus number \u2014 saying otherwise would be the fabrication this course forbids.\n", "if len(Xtr) > 120_000:\n", " Xtr, _, ytr, _ = train_test_split(Xtr, ytr, train_size=120_000, random_state=RANDOM_STATE,\n", " stratify=ytr) # genuinely stratified, not random\n", "MAJORITY_BASELINE = max(np.mean(yte), 1 - np.mean(yte)) # accuracy of 'always predict majority'\n", "print(f'materialized {N_MATERIALIZED:,} rows | trained on {len(Xtr):,} (stratified subsample) | '\n", " f'held-out {len(yte):,}')\n", "print(f'MAJORITY-CLASS BASELINE accuracy = {MAJORITY_BASELINE:.4f} '\n", " f'(any model must beat THIS, not 0.5, to be interesting)')\n", "\n", "models = { # four standard, diverse learners\n", " 'LogisticRegression': make_pipeline(StandardScaler(), LogisticRegression(max_iter=300)), # scaled!\n", " 'RandomForest': RandomForestClassifier(n_estimators=60, n_jobs=-1, random_state=RANDOM_STATE),\n", " 'XGBoost': xgb.XGBClassifier(n_estimators=80, max_depth=6, tree_method='hist', n_jobs=-1,\n", " eval_metric='logloss', random_state=RANDOM_STATE),\n", " 'LightGBM': lgb.LGBMClassifier(n_estimators=80, n_jobs=-1, verbose=-1, random_state=RANDOM_STATE),\n", "}\n", "rows, fitted = [], {}\n", "for name, m in models.items(): # fit + score each model\n", " t = time.perf_counter(); m.fit(Xtr, ytr); fitted[name] = m\n", " p = m.predict_proba(Xte)[:, 1] # positive-class probability on held-out\n", " rows.append({'model': name, 'accuracy': round(accuracy_score(yte, (p>0.5).astype(int)), 6),\n", " 'roc_auc': round(roc_auc_score(yte, p), 6), # 6 dp: a 1.000000 is a red flag, not a win\n", " 'train_s': round(time.perf_counter()-t, 1)})\n", "rows.append({'model': 'MajorityBaseline', 'accuracy': round(MAJORITY_BASELINE, 4),\n", " 'roc_auc': 0.5, 'train_s': 0.0}) # show the baseline IN the ranking table\n", "comparison = pd.DataFrame(rows).sort_values('roc_auc', ascending=False).reset_index(drop=True)\n", "_ranked = comparison[comparison.model != 'MajorityBaseline']\n", "best_name = _ranked.iloc[0]['model']; best = fitted[best_name] # winner by ROC-AUC (excl. baseline)\n", "print('best model:', best_name); comparison" ] }, { "cell_type": "markdown", "id": "3140800d", "metadata": {}, "source": [ "## 9. Results\n", "\n", "Diagnostics for the winning model, including **per-group recall**.\n", "\n", "The grouping comes from whatever the loader put in `family`. It is *not* always an attack taxonomy. On the intrusion corpora it is the attack family. On the fraud and malware corpora it is a transaction type, a merchant category or a malware category. On binary corpora it collapses to the positive class.\n", "\n", "Read it accordingly. Where the groups are genuinely rare classes, they reveal whether detection is real. The dominant flood classes do not.\n", "\n", "Here the groups are HDFS anomaly `Type` codes. The confusion-matrix ticks and the printed false-positive line still read `benign` and `attack`. Read them as **Success** and **Fail** \u2014 see section 7.\n", "\n", "**Coverage limit on the recall panel:** a group is scored only when it has at least five held-out positives (`if cnt < 5: continue`). Types under that floor are skipped silently. They reach neither the chart nor the printed list, so their absence there is not evidence of detection." ] }, { "cell_type": "code", "execution_count": 6, "id": "efa288e1", "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "operational FALSE-POSITIVE RATE @0.5 = 0.0001 (19 benign flagged of 139,556)\n", "worst per-family recalls: {'type_21.0': 0.995, 'type_4.0': 0.997, 'type_0.0': 0.997, 'type_5.0': 1.0, 'type_31.0': 1.0, 'type_3.0': 1.0}\n" ] } ], "source": [ "# --- Results for the best model: confusion, ROC, PR, importances, per-family recall ---\n", "from sklearn.metrics import confusion_matrix, roc_curve, precision_recall_curve, recall_score\n", "pb = best.predict_proba(Xte)[:, 1]; pred = (pb > 0.5).astype(int)\n", "fig, ax = plt.subplots(1, 3, figsize=(15, 4))\n", "# (1) confusion matrix\n", "cm = confusion_matrix(yte, pred); ax[0].imshow(cm, cmap='Blues')\n", "ax[0].set_title(f'{best_name}: confusion'); ax[0].set_xticks([0,1]); ax[0].set_yticks([0,1])\n", "ax[0].set_xticklabels([NEG_WORD,POS_WORD]); ax[0].set_yticklabels([NEG_WORD,POS_WORD])\n", "for (i,j),v in np.ndenumerate(cm): ax[0].text(j,i,f'{v:,}',ha='center',va='center')\n", "# (2) ROC and PR curves\n", "fpr,tpr,_ = roc_curve(yte, pb); prec,rec,_ = precision_recall_curve(yte, pb)\n", "ax[1].plot(fpr,tpr,color='#264653'); ax[1].plot([0,1],[0,1],'--',c='grey')\n", "ax[1].set_title(f'ROC (AUC={roc_auc_score(yte,pb):.4f})'); ax[1].set_xlabel('FPR'); ax[1].set_ylabel('TPR')\n", "ax[2].plot(rec,prec,color='#e76f51'); ax[2].set_title('Precision-Recall'); ax[2].set_xlabel('recall'); ax[2].set_ylabel('precision')\n", "plt.tight_layout(); plt.show()\n", "\n", "# (3) feature importances + (4) per-attack-family recall\n", "fig, ax = plt.subplots(1, 2, figsize=(13, 5))\n", "imp, names = None, feat # importances, robust to the scaled-LR pipeline\n", "if hasattr(best, 'feature_importances_'): # tree models\n", " imp = best.feature_importances_; names = list(getattr(best, 'feature_names_in_', feat))[:len(imp)]\n", "elif hasattr(best, 'named_steps') and 'logisticregression' in getattr(best, 'named_steps', {}):\n", " imp = np.abs(best.named_steps['logisticregression'].coef_[0]); names = feat # LR pipeline\n", "elif hasattr(best, 'coef_'):\n", " imp = np.abs(best.coef_[0]); names = feat\n", "if imp is not None:\n", " pd.Series(imp, index=names[:len(imp)]).sort_values().tail(12).plot.barh(ax=ax[0], color='#264653')\n", "ax[0].set_title(f'{best_name}: top importances / |coef|')\n", "# Per-family recall, WORST-first so rare, hard classes are visible, not just the dominant floods.\n", "fam_te = df.loc[Xte.index, 'family']\n", "fr = {}\n", "for fam, cnt in fam_te[yte==1].value_counts().items():\n", " if cnt < 5: continue # need a few positives for a meaningful recall\n", " mask = (fam_te==fam).to_numpy(); fr[fam] = recall_score(yte[mask], pred[mask], zero_division=0)\n", "srt = pd.Series(fr).sort_values()\n", "show = pd.concat([srt.head(9), srt.tail(3)]) if len(srt) > 12 else srt # worst 9 + best 3\n", "show = show[~show.index.duplicated()]\n", "show.plot.barh(ax=ax[1], color=['#e76f51' if v < 0.5 else '#2a9d8f' for v in show]); ax[1].set_xlim(0,1)\n", "ax[1].set_title('Per-family recall (worst first; red < 0.5)')\n", "plt.tight_layout(); plt.show()\n", "# Operational numbers, not just figures: false-positive rate and the worst per-family recalls.\n", "tn, fp = int(cm[0,0]), int(cm[0,1])\n", "fpr_op = fp/(fp+tn) if (fp+tn) > 0 else float('nan') # benign wrongly flagged @0.5\n", "print(f'operational FALSE-POSITIVE RATE @0.5 = {fpr_op:.4f} ({fp:,} benign flagged of {fp+tn:,})')\n", "print('worst per-family recalls:', {k: round(v, 3) for k, v in srt.head(6).items()})" ] }, { "cell_type": "markdown", "id": "1bde3087", "metadata": {}, "source": [ "## 10. Validity audit \u2014 is the score real?\n", "\n", "Three diagnostics. **(a)** How well can the *single best feature*, alone, separate the classes? A near-1.0 single-feature AUC means that feature is *near-sufficient* \u2014 a shortcut (which may be legitimate signal or an artifact), not the same as target leakage. **(b)** The exact-duplicate row rate. **(c)** The **train/test exact-row contamination** \u2014 the fraction of held-out rows that are duplicates of training rows, which is what actually inflates a held-out score. The trust grade is the *worse* of the single-feature and contamination concerns." ] }, { "cell_type": "code", "execution_count": 7, "id": "5f67e6d0", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "best single-feature AUC = 0.6896 (feature: E9)\n", " note: a near-1.0 single-feature AUC means this feature is *near-sufficient* (a shortcut),\n", " which may be legitimate signal OR an artifact \u2014 it is NOT the same as target leakage.\n", "exact-duplicate row rate (whole corpus) = 0.999\n", "TRAIN/TEST exact-row contamination = 0.999 (single-feat grade A, contam grade F)\n", "==> data trust grade: F (worse of the two; F = shortcut and/or heavy contamination)\n" ] }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# --- Validity audit: is the score real detection, or a data shortcut? ---\n", "from sklearn.metrics import roc_auc_score\n", "samp = X.sample(min(60_000, len(X)), random_state=1); ysamp = y[samp.index]\n", "aucs = {}\n", "for c in feat: # AUC of EACH feature alone\n", " col = samp[c].to_numpy(float)\n", " if col.std()==0: continue\n", " a = roc_auc_score(ysamp, col); aucs[c] = max(a, 1-a) # direction-agnostic\n", "best_auc = max(aucs.values()); best_col = max(aucs, key=aucs.get)\n", "dup_rate = 1 - X.drop_duplicates().shape[0]/len(X) # exact-duplicate feature rows (whole set)\n", "# The statistic that actually inflates a held-out score is TRAIN/TEST CONTAMINATION: how many test\n", "# rows are exact duplicates of a training row. Measure it directly on the split used above.\n", "_trkeys = set(map(tuple, np.round(Xtr.to_numpy(), 6)))\n", "_te = np.round(Xte.to_numpy(), 6)[:50_000]\n", "contam = float(np.mean([tuple(r) in _trkeys for r in _te])) # fraction of test rows seen in train\n", "# Trust grade reflects BOTH failure modes and takes the WORSE of the two: a near-perfect single\n", "# feature (shortcut) OR heavy train/test contamination each independently invalidate the headline.\n", "_ga = 'F' if best_auc>=0.999 else 'D' if best_auc>=0.99 else 'C' if best_auc>=0.95 else 'B' if best_auc>=0.85 else 'A'\n", "_gc = 'F' if contam>=0.5 else 'D' if contam>=0.3 else 'C' if contam>=0.15 else 'B' if contam>=0.05 else 'A'\n", "grade = max(_ga, _gc) # 'max' letter = worse grade (A best, F worst)\n", "print(f'best single-feature AUC = {best_auc:.4f} (feature: {best_col})')\n", "print(f' note: a near-1.0 single-feature AUC means this feature is *near-sufficient* (a shortcut),\\n'\n", " f' which may be legitimate signal OR an artifact \u2014 it is NOT the same as target leakage.')\n", "print(f'exact-duplicate row rate (whole corpus) = {dup_rate:.3f}')\n", "print(f'TRAIN/TEST exact-row contamination = {contam:.3f} (single-feat grade {_ga}, contam grade {_gc})')\n", "print(f'==> data trust grade: {grade} (worse of the two; F = shortcut and/or heavy contamination)')\n", "s = pd.Series(aucs).sort_values().tail(15)\n", "fig, ax = plt.subplots(figsize=(8,5))\n", "s.plot.barh(ax=ax, color=['#e76f51' if v>=0.99 else '#457b9d' for v in s]); ax.axvline(0.5,ls='--',c='grey')\n", "ax.set_xlim(0.5,1.0); ax.set_title('Single-feature ROC-AUC (red = near-perfect shortcut)'); ax.set_xlabel('AUC alone')\n", "plt.tight_layout(); plt.show()" ] }, { "cell_type": "markdown", "id": "c8d2b138", "metadata": {}, "source": [ "## 11. Ablation \u2014 does the headline survive removing the artifacts?\n", "\n", "Narrating a shortcut is not enough. We *retrain the winning model* after (1) de-duplicating the corpus (removing the train/test contamination) and (2) dropping the single strongest feature. We report the held-out AUC each time. **Read the result honestly, both ways:** if the AUC **collapses**, the headline was a contamination/shortcut artifact. If it **barely moves** \u2014 common on *simulated* corpora \u2014 that is **not vindication**. It means the classes are separable by *many* redundant features, because the two class distributions barely overlap. That is its own generation artifact. The numbers below decide which story is true here, not the prose." ] }, { "cell_type": "code", "execution_count": 8, "id": "d0017f8a", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Ablation \u2014 how much of the headline survives once each artifact is removed:\n" ] }, { "data": { "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", "
settingheld_out_auc
0headline (as-is)0.999983
1de-duplicated (100% rows removed)0.983574
2shortcut feature dropped (E9)0.999979
\n", "
" ], "text/plain": [ " setting held_out_auc\n", "0 headline (as-is) 0.999983\n", "1 de-duplicated (100% rows removed) 0.983574\n", "2 shortcut feature dropped (E9) 0.999979" ] }, "execution_count": 8, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# --- Ablation: SHOW the inflation empirically, don't just narrate it ---\n", "from sklearn.base import clone\n", "def _retrain_auc(Xa, ya): # re-split, stratified-subsample, refit best family\n", " xtr, xte, ytr2, yte2 = train_test_split(Xa, ya, test_size=0.25, random_state=RANDOM_STATE, stratify=ya)\n", " if len(xtr) > 120_000:\n", " xtr, _, ytr2, _ = train_test_split(xtr, ytr2, train_size=120_000, random_state=RANDOM_STATE, stratify=ytr2)\n", " m = clone(best); m.fit(xtr, ytr2)\n", " return roc_auc_score(yte2, m.predict_proba(xte)[:, 1])\n", "base_auc = roc_auc_score(yte, best.predict_proba(Xte)[:, 1]) # (0) the headline held-out AUC\n", "Xdd = X.drop_duplicates(); ydd = y[Xdd.index] # (1) de-duplicated corpus\n", "auc_dedup = _retrain_auc(Xdd, ydd)\n", "auc_noshort = _retrain_auc(X.drop(columns=[best_col]), y) if best_col in X.columns else base_auc # (2) drop shortcut\n", "ablation = pd.DataFrame([\n", " {'setting': 'headline (as-is)', 'held_out_auc': round(base_auc, 6)},\n", " {'setting': f'de-duplicated ({1-len(Xdd)/len(X):.0%} rows removed)', 'held_out_auc': round(auc_dedup, 6)},\n", " {'setting': f'shortcut feature dropped ({best_col})', 'held_out_auc': round(auc_noshort, 6)},\n", "])\n", "print('Ablation \u2014 how much of the headline survives once each artifact is removed:')\n", "ablation" ] }, { "cell_type": "markdown", "id": "576b1844", "metadata": {}, "source": [ "## 12. Reproducibility & robustness\n", "\n", "**Scope of the CV number below, stated plainly:** the folds are not the corpus. The cell slices the head off the training subsample (`cvX, cvy = Xtr.iloc[:40_000], ytr[:40_000]`) and cross-validates on that slice alone. `Xtr` is itself the stratified subsample whose size the model-comparison cell printed. So the CV figure describes a slice of a subsample, two steps removed from the 575,061 sessions loaded.\n", "\n", "The slice is also never de-duplicated. Section 10 prints an exact-duplicate row rate of 0.999 for the whole corpus, so duplicate rows very likely straddle these three folds as well. The notebook never measures that, so treat it as a mechanism to check rather than a printed result. One-line check: `Xtr.iloc[:40_000].duplicated().mean()`.\n", "\n", "Read the CV mean as a stability check on one split. It is not evidence that contamination has been ruled out. The de-duplicated row of the section 11 ablation is the number that speaks to contamination." ] }, { "cell_type": "code", "execution_count": 9, "id": "ec7d8d36", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "seed=0 | numpy 2.3.5 | sklearn 1.9.0 | xgboost 1.6.2 | lightgbm 4.7.0\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "RandomForest 3-fold CV ROC-AUC = 0.9996 +/- 0.0006 (mean +/- std across 3 stratified folds; a small std means a stable estimate on this split)\n" ] } ], "source": [ "# --- Reproducibility & robustness ---\n", "import sklearn\n", "from sklearn.model_selection import StratifiedKFold, cross_val_score\n", "print(f'seed={RANDOM_STATE} | numpy {np.__version__} | sklearn {sklearn.__version__} | '\n", " f'xgboost {xgb.__version__} | lightgbm {lgb.__version__}')\n", "# 3-fold cross-validated ROC-AUC of the winning model (fresh clone, bounded subsample) -> mean +/- std.\n", "from sklearn.base import clone\n", "cvX, cvy = Xtr.iloc[:40_000], ytr[:40_000]\n", "def _auc_scorer(est, Xv, yv): # robust to xgboost's 2-col predict_proba\n", " p = est.predict_proba(Xv)\n", " p = p[:, 1] if getattr(p, 'ndim', 1) == 2 else p\n", " return roc_auc_score(yv, p)\n", "try:\n", " cv = cross_val_score(clone(best), cvX, cvy,\n", " cv=StratifiedKFold(3, shuffle=True, random_state=RANDOM_STATE),\n", " scoring=_auc_scorer, error_score='raise')\n", " assert np.all(np.isfinite(cv)), 'non-finite CV folds' # FAIL CLOSED: never narrate a NaN as evidence\n", " print(f'{best_name} 3-fold CV ROC-AUC = {cv.mean():.4f} +/- {cv.std():.4f} '\n", " f'(mean +/- std across 3 stratified folds; a small std means a stable estimate on this split)')\n", "except Exception as e:\n", " print(f'CV UNAVAILABLE ({type(e).__name__}: {str(e)[:60]}); rely on the single held-out AUC above \u2014 '\n", " f'we do NOT report a CV number we could not compute')" ] }, { "cell_type": "markdown", "id": "80147cc0", "metadata": {}, "source": [ "## 13. Scientific conclusion\n", "\n", "Event-count vectors separate normal from anomalous HDFS sessions very sharply. Per-anomaly-type recall is high for every type the notebook actually **scored** \u2014 which is not the same as every failure type. The recall cell skips any type with fewer than five held-out positives (`if cnt < 5: continue`), and it never prints how many types that removes. The anomaly `Type` codes have a long tail, so treat the skipped set as non-empty until you check it: `df.loc[df.y==1,'family'].value_counts().tail(15)`. The printed recalls cannot be read as full coverage. Two further caveats reshape the headline. **(1) Contamination is real here.** HDFS count vectors are so redundant that nearly every session shares an identical vector with another. A random split therefore leaks (grade F). Unlike some datasets, it *matters*: de-duplicating drops the AUC from ~1.0 to ~0.98. **(2) Counts ignore event order**, so ordering-only anomalies are invisible to this representation \u2014 a sequence model (DeepLog) is the honest next step. With ~3% anomalies, per-type recall and the false-positive rate \u2014 not accuracy \u2014 remain the metrics that matter (Xu et al., 2009; Sommer & Paxson, 2010).\n", "\n", "**Validity ledger \u2014 read the headline against these printed numbers:** Majority-class baseline **accuracy**: **0.9707**. The accuracy column must clear that bar to mean anything. For ROC-AUC the trivial baseline is 0.5, not that figure. Winning learner: **RandomForest**. Its 3-fold CV ROC-AUC is **0.9996** \u2014 but read the scope in section 12. Those folds run on the head slice `Xtr.iloc[:40_000]` of the training subsample, not on the corpus, and that slice is not de-duplicated. The figure measures split stability, not freedom from contamination. Strongest *single* feature: `E9` at AUC **0.6896**. The ablation refutes a single-feature story. Dropping that feature barely moves the AUC: **0.999983 \u2192 0.999979**. So the separability is **multi-feature**. That reflects how this corpus was generated, not one leaky column. *(Read that table row with care. It rounds to whole percents, so a duplicate rate of **0.9990** prints as \u2018100% rows removed\u2019. It is not literally 100%. A full removal would leave nothing to refit on.)* Data-trust grade: **F**. It is the worse of two independent sub-checks. Single-feature AUC 0.6896 scores **A**. Train/test exact-row overlap 0.999 scores **F**. The overlap check drives the grade, not the single-feature check. That says the split leaks, not that features are clean; the single-feature check separately scores A. On this A-best / F-worst scale, a D or F means the headline is optimistic. Treat it as a benchmark number, not a deployment estimate. Operational false-positive rate at threshold 0.5: **0.0001**. Worst per-group recalls, exactly as printed: {`type_21.0`: 0.995, `type_4.0`: 0.997, `type_0.0`: 0.997, `type_5.0`: 1.0, `type_31.0`: 1.0, `type_3.0`: 1.0}. The weakest **scored** group sits at **0.995**, which is where measured detection is thinnest. Groups under the five-positive floor are absent from that list, so it bounds the scored groups only. **Disclosed limitation:** categorical columns are integer-encoded before the split. The encoder therefore sees the test set's category values. On an all-numeric corpus that step is a no-op. The mapping never consults the label, so no *label* information leaks. It is still transductive. A deployed system would need an unseen-category bucket. **How the audit numbers are computed:** overlap is measured on the first 50,000 held-out rows, so read it as a sampled estimate. Each ablation re-splits and refits, so tiny differences are re-split noise. The de-duplication variant keeps the first label when a feature vector appears twice. **Scope:** the split is random, not temporal or entity-grouped. Every number above therefore measures in-distribution separability only." ] }, { "cell_type": "markdown", "id": "1f691013", "metadata": {}, "source": [ "## References\n", "\n", "1. Xu, W., Huang, L., Fox, A., Patterson, D. & Jordan, M.I. (2009). Detecting Large-Scale System Problems by Mining Console Logs. *ACM SOSP*.\n", "2. Zhu, J., He, S., He, P., Liu, J. & Lyu, M.R. (2023). Loghub: A Large Collection of System Log Datasets for AI-driven Log Analytics. *IEEE Int. Symposium on Software Reliability Engineering (ISSRE)*, 355-366 (preprint arXiv:2008.06448, 2020).\n", "3. Du, M., Li, F., Zheng, G. & Srikumar, V. (2017). DeepLog: Anomaly Detection and Diagnosis from System Logs through Deep Learning. *ACM CCS*.\n", "4. Sommer, R. & Paxson, V. (2010). Outside the Closed World: On Using Machine Learning for Network Intrusion Detection. *IEEE S&P*.\n", "5. He, P., Zhu, J., Zheng, Z. & Lyu, M.R. (2017). Drain: An Online Log Parsing Approach with Fixed Depth Tree. *IEEE International Conference on Web Services (ICWS)*, 33-40. Cited as a background parsing method only; it did not produce this notebook's event counts." ] } ], "metadata": { "kernelspec": { "display_name": "seas8414-phase-notebooks", "language": "python", "name": "seas8414-phase-notebooks" } }, "nbformat": 4, "nbformat_minor": 5 }