Estudio comparativo de herramientas de procesamiento de Lenguaje Natural Open Source y Software propietario para el desarrollo de aplicaciones basadas en inteligencia artificial.

The goal of this comparative study is to identify the main Open-Source and Proprietary Natural Language Processing (NLP) tools and to analyze one Open-Source NLP tool and one Proprietary tool in terms of their functionality, efficiency, price, and ease of implementation for the development of artificial intelligence-based applications. To this end, a research was carried out on Natural Language Processing and its main applications. Regarding Open-Source tools, Spacy, NLTK, Gensim, and Spark NLP were selected highlighting their main features. In the same way, this process was carried out with Google Natural Language API, IBM Watson Natural Language Understanding, Amazon Comprehend, and Azure AI API, which are proprietary tools. The advantages and disadvantages of using these two types of tools were compared, thus facilitating informed decision-making. In terms of practical analysis two tools were selected: Google Natural Language API with the Google Cloud platform and the Google Colab platform which was used to run Spacy. It is concluded that Open-Source and proprietary tools offer excellent results in terms of text analysis. The selection of one tool or another should be based on the specific needs and characteristics of each project, as well as the available budget as a company or developer.

• Glavni autor: Angamarca Peña, Klever Eduardo (author)
• Format: bachelorThesis
• Izdano: 2023
• Teme: NLP; Spacy; Open Source; Google API; Google Cloud; AI
• Online pristup: http://dspace.utb.edu.ec/handle/49000/15028