Integrating Autoencoders with Local Outlier Factor and Isolation Forest for Effective Fraud Detection in Imbalanced Datasets
Bibliographic record
Abstract
Abstract: It is highly challenging to detect fraudulent transactions with extant imbalances in available datasets where fraudulent cases make up a minor percentage of total transactions. This work presents a novel hybrid anomaly detection framework that integrates Autoencoders for efficient dimensionality reduction with LOF and Isolation Forest algorithms to detect anomalies for accurate fraud detection. We make use of the very standard dataset, namely Credit Card Fraud Detection Dataset [7], that has 284,807 transactions of which only 492 are classified as fraudulent. We apply Synthetic Minority Over-Sampling Technique to balance the dataset for optimizing the model’s performance. The results show that although LOF is challenging in terms of precision, it exhibits significant increases in recall with the proper adjustment of the contamination parameter and utilization of SMOTE. In comparison, Isolation Forest algorithm works excellently in terms of recall where it detects 81% frauds but degrades slightly in terms of precision after using SMOTE. The two techniques here have trade-offs between precision and recall, hence indicating a scope for further optimization. Both LOF and Isolation Forest significantly contribute in detecting anomalies in imbalanced datasets, and though Isolation Forest has a higher efficiency ratio compared to LOF in fraud transaction detection, our results confirm that indeed using Autoencoders for the extraction of features and advanced anomaly detection techniques have a synergistic effect in fraud detection applications, particularly in big class imbalance scenarios. Future research would include other oversampling techniques along with fine-tuning the parameter settings to have a better balance between precision and recall.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".