KAI-KratosAI commited on
Commit
a322250
·
verified ·
1 Parent(s): 41ad434

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +77 -3
README.md CHANGED
@@ -1,3 +1,77 @@
1
- ---
2
- license: cc-by-4.0
3
- ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: cc-by-4.0
3
+ pretty_name: 'Bengali Documents Dataset (PDF)'
4
+ language:
5
+ - bn
6
+ tags:
7
+ - pdf
8
+ - bengali
9
+ - document-understanding
10
+ - text-recognition
11
+ - ocr
12
+ - ai-research
13
+ - computer-vision
14
+ - text-extraction
15
+ task_categories:
16
+ - document-classification
17
+ - text-recognition
18
+ size_categories:
19
+ - 1K<n<10K
20
+ ---
21
+
22
+ # Bengali Documents Dataset (PDF)
23
+ *This dataset contains a curated collection of Bengali-language documents in PDF format. The corpus includes educational materials, books, reports, articles, and public-domain texts written in Bengali script. It is designed to support AI research in OCR, document understanding, and text extraction for Indic languages.*
24
+
25
+ ## Contact
26
+ For queries or collaborations related to this dataset, contact:
27
+ - anoushka@kgen.io
28
+ - abhishek.vadapalli@kgen.io
29
+
30
+ ## Supported Tasks
31
+
32
+ - **Task Categories**:
33
+ - Document Classification
34
+ - OCR and Text Recognition
35
+ - Layout and Structure Analysis
36
+ - Language Modeling for Bengali
37
+
38
+ - **Supported Tasks**:
39
+ - Extraction of Bengali text from PDF documents
40
+ - Classification by topic or domain (education, literature, news, government)
41
+ - OCR for printed Bengali script and mixed-language layouts
42
+ - Benchmarking Bengali-language document processing models
43
+
44
+ ## Languages
45
+
46
+ - **Primary Language**: Bengali
47
+ - **Secondary Presence**: English terms, numerals, and section headers (common in bilingual publications)
48
+
49
+ ## Dataset Creation
50
+
51
+ ### Curation Rationale
52
+ The dataset was curated to advance AI systems that process Bengali-language documents, enhance OCR quality for Indic scripts, and enable structured understanding of mixed-script PDFs.
53
+
54
+ ### Source Data
55
+ - **Contributors**: Public archives, open-access Bengali libraries, educational repositories, and volunteer submissions.
56
+ - **Collection Process**: Documents were compiled from freely available PDF sources with permissible open licenses.
57
+
58
+ ### Other Known Limitations
59
+ - **Bias**: Focused more on educational and literary content; limited representation of informal or handwritten material
60
+ - **Format Variation**: Some PDFs include embedded images or scanned pages
61
+ - **Dialect Coverage**: Primarily Standard Bengali (বাংলা), limited regional dialects
62
+
63
+ ## Intended Uses
64
+
65
+ ### ✅ Direct Use
66
+ - Training OCR and NLP systems for Bengali document understanding
67
+ - Research on multilingual document parsing and Indic language digitization
68
+ - Educational and academic AI research for Bengali text extraction
69
+
70
+ ### ❌ Out-of-Scope Use
71
+ - Extracting or profiling personal data from documents
72
+ - Commercial reuse of copyrighted content
73
+ - Use in surveillance, handwriting identification, or behavioral analysis
74
+
75
+ ## License
76
+
77
+ CC BY 4.0