create README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,82 @@
|
|
| 1 |
-
---
|
| 2 |
-
license: apache-2.0
|
| 3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
task_categories:
|
| 4 |
+
- table-question-answering
|
| 5 |
+
- question-answering
|
| 6 |
+
language:
|
| 7 |
+
- en
|
| 8 |
+
tags:
|
| 9 |
+
- finance
|
| 10 |
+
- legal
|
| 11 |
+
size_categories:
|
| 12 |
+
- 100K<n<1M
|
| 13 |
+
---
|
| 14 |
+
# SEC 10-X Filings Dataset
|
| 15 |
+
|
| 16 |
+
This dataset contains processed SEC 10-X (10-K, 10-Q) filings, focusing on Risk Factors and Management Discussion & Analysis (MD&A) sections from corporate financial reports from 1993-2023.
|
| 17 |
+
|
| 18 |
+
🔗 Original Dataset: [[SEC-EDGAR-10X]](https://sraf.nd.edu/sec-edgar-data/cleaned-10x-files/) contains stripped down versions of the original filings, details about which can be found [here](https://sraf.nd.edu/sec-edgar-data/cleaned-10x-files/10x-stage-one-parsing-documentation/).
|
| 19 |
+
|
| 20 |
+
This dataset is a further cleaned tabulated version of the original stripped down version making it more suitable for training tasks.
|
| 21 |
+
|
| 22 |
+
Note: Documents with no match to risk factors or MD&A have been filtered out.
|
| 23 |
+
|
| 24 |
+
## Dataset Description
|
| 25 |
+
|
| 26 |
+
### Overview
|
| 27 |
+
|
| 28 |
+
This dataset is derived from SEC 10-X filings and provides structured access to two critical sections of corporate financial reports:
|
| 29 |
+
- Risk Factors (Item 1A)
|
| 30 |
+
- Management's Discussion and Analysis (Item 7)
|
| 31 |
+
|
| 32 |
+
The data has been processed to enable natural language processing and financial analysis tasks.
|
| 33 |
+
|
| 34 |
+
### Processing Methodology
|
| 35 |
+
|
| 36 |
+
The dataset was created using a parsing pipeline that addresses several key considerations:
|
| 37 |
+
|
| 38 |
+
1. **Risk Factors Extraction**
|
| 39 |
+
- Multiple instances of risk factors may appear in a single filing
|
| 40 |
+
- The parser identifies all instances and selects the most comprehensive section
|
| 41 |
+
- When multiple valid sections are found, the longest section is retained to ensure completeness
|
| 42 |
+
|
| 43 |
+
2. **MD&A Extraction**
|
| 44 |
+
- Management Discussion & Analysis sections are identified and extracted
|
| 45 |
+
- Multiple MD&A sections within a filing are concatenated to preserve all relevant information
|
| 46 |
+
|
| 47 |
+
3. **Data Organization**
|
| 48 |
+
- Filings are grouped by CIK (Central Index Key) and filing date
|
| 49 |
+
- Multiple sections from the same filing are combined
|
| 50 |
+
- Empty or invalid entries are filtered out
|
| 51 |
+
|
| 52 |
+
### Data Format
|
| 53 |
+
|
| 54 |
+
The dataset is stored in Parquet format with the following schema:
|
| 55 |
+
```python
|
| 56 |
+
{
|
| 57 |
+
'CSI': 'string', # Central Index Key (CIK)
|
| 58 |
+
'FILE_DATE': 'string', # Filing date in YYYYMMDD format
|
| 59 |
+
'RISK_FACTOR': 'string', # Extracted Risk Factors section
|
| 60 |
+
'MD&A': 'string' # Extracted Management Discussion & Analysis section
|
| 61 |
+
}
|
| 62 |
+
```
|
| 63 |
+
## Usage
|
| 64 |
+
|
| 65 |
+
### Loading the Dataset
|
| 66 |
+
```python
|
| 67 |
+
import datasets
|
| 68 |
+
dataset = datasets.load_dataset("theaayushbajaj/10-X-raw-v1", split="train")
|
| 69 |
+
```
|
| 70 |
+
|
| 71 |
+
### Example Applications
|
| 72 |
+
|
| 73 |
+
- Risk analysis and classification
|
| 74 |
+
- Temporal analysis of corporate risk factors
|
| 75 |
+
- Business strategy analysis through MD&A
|
| 76 |
+
- Corporate disclosure analysis
|
| 77 |
+
- Financial sentiment analysis
|
| 78 |
+
|
| 79 |
+
## Acknowledgments
|
| 80 |
+
|
| 81 |
+
- Securities and Exchange Commission (SEC) for providing access to the original filings
|
| 82 |
+
- University of Notre Dame for compiled versions
|