Textar

Head of Cabinet of Ministers

Textar is a set of Python algorithms designed to consolidate natural language processing (NLP) tools used by the National Directorate of Public Information and Data of Argentina. This tool makes it easy to search and categorize questions for government reporting. Developed to assist in the preparation of reports and resolve specific questions, Textar improves efficiency in the management of government questions, avoiding duplication and categorizing by topic. These algorithms are a fundamental part of an internal interface that optimizes the integration and communication of information.

Textar solves the problem of managing large volumes of questions and data within government offices, simplifying the search and categorization of information, which is crucial for efficient reporting and internal communication.

Text analysis and classification. Retrieval of similar texts. Utilization of machine learning algorithms. Integration with Python data science stack.

Built on Python as the core language, leveraging its versatility and efficiency. Integrates open-source libraries such as pandas, numpy, scikit-learn, and scipy for data analysis, machine learning, and scientific computations. Adopts the MIT license to promote openness and collaboration in development. Facilitates interoperability and the use of open data formats to maximize accessibility and information reuse.