UNAM Thesis Texts from Five Academic Programs

Filled with 1,000 Spanish-language theses from the National Autonomous University of Mexico (UNAM), with 200 theses from each of 5 academic programs. It is intended for text classification tasks and is monolingual in Spanish.

Use Cases

• Perform text classification on thesis documents to categorize them by their academic program.
• Analyze linguistic patterns across the 5 distinct academic programs represented in the collection.
• Train a language model on a corpus of 1,000 formal academic documents in Spanish.

Strengths

• Contains 1,000 individual thesis documents.
• Provides a balanced sample of 200 theses from each of 5 academic programs.
• Specifically curated for the text classification task.

Limitations

• Limited to 5 academic programs, not representing the full scope of UNAM's offerings.
• Sample size of 1,000 theses may be insufficient for training large language models.
• Potential for class imbalance if theses are not evenly distributed across all relevant sub-topics within a program.

Provenance

Source

National Autonomous University of Mexico (UNAM).

Collection Method

Crowdsourced collection, as indicated by dataset tags.

Freshness

Last updated on October 11, 2023.

Geography

Mexico, specifically from the National Autonomous University of Mexico (UNAM).