Research on intelligent financial statement analysis and anomaly identification technology based on machine learning
Bibliographic record
Abstract
In order to accurately assess the financial status of a company and identify potential anomalies, this paper first implements unsupervised classification of financial transaction data based on Support Vector Machines, which automatically classifies the data into normal and abnormal categories.Histograms are introduced in combination with LightGBM to quickly fuse data from multiple sources.The most suitable first layer is selected by different algorithms, and the outputs of these algorithms are combined with industry-wide common abnormal features as inputs for LightGBM's second layer identification.With this two-layer structure, the model not only takes into account the industry characteristics, but also the common anomaly features.Empirical results show that in the accuracy of smart financial statement generation, the sensitivity of this paper's model iterates to 99.99% at 41.25% specificity, and the accuracy of this paper's model is as high as 0.98 when dealing with financial private information, macroeconomic, and market information.In the identification of financial transaction anomalies, the number of anomalous weeks is identified to be 24, 29, 34, and 36, and the fusion of multi-source data effectively identifies the large amount of financial transactions, fluctuating transactions and other suspicious abnormal transactions.
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 machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.005 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.003 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.002 | 0.004 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 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 source (direct Gemma or distilled Codex), 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".