The continuous increase of digital documents on the web creates the need to search for information patterns that allow the categorization of organizational documents to generate knowledge in an institution. An Artificial Intelligence technique for this purpose is text classification, it for its application uses labels (previously categorized documents) with supervised (with labels) or unsupervised (without labels) training models. Both traditional models with their advantages and disadvantages have been joined into semi-supervised models that extract the best qualities of each one, however, the labeling process involves resources and time that try to be optimized to improve classification accuracy. An analysis of the different semi-supervised models would show us the advantages of their training and the way how the structure of each of them affects the accuracy of their classification. In the present study, a classification structure of the semi-supervised models in the classification of documents is proposed to analyze their qualities and categorization process, through an SLR (Revision of systematic literature) that extracts performance metrics from the identified studies to perform a meta-analysis through forest plots. To define the search strategy for studies, the PICOC (Population, Intervention, Comparison, Outputs, Context) method has been used, it is supported by the research question defines a search string, which has allowed the collection of 228 research, these are filtered with the PRISMA declaration method and the determination of exclusion criteria, in this way 35 researches are selected for the present study. The analysis of the selected studies identifies a structure for the different semi-supervised learning models, and a scheme of their work process is obtained, it has been used to extract advantages, disadvantages, and performance metrics. Through a meta-analysis with forest diagrams, the classification accuracy performance of the researches in each learning model is evaluated, determining as results that regardless of the characteristics of its process, active learning (0.89) and assembled learning (0.83) present the best performance levels.
1 April 2014 in Ingeniare. Revista chilena de ingeniería
In the context of the digital era, our lives have become more dynamic, and the development of advanced technologies is transforming the way civil law relationships are handled. Today, legal professionals commonly use software that accelerates daily processes through the application of artificial intelligence technologies. Legal Information Retrieval is a significant and challenging field of AI that focuses on finding and analyzing legal norms and documents relevant to a user’s information needs. The objective of this article is to identify and synthesize the main approaches, trends, and advances in the application of AI in Legal Information Retrieval. Through a review of recent research, this study aims to provide a clear overview of the strategies used, methodologies employed, and emerging areas of focus. To achieve these objectives, exhaustive search methods were applied to academic databases and repositories, selecting relevant studies published over the past twelve years. In the first stage, 2307 articles were found. Then, in the second stage, 354 technical articles were selected after applying inclusion/exclusion criteria. Finally, after a third round of filtering, 18 articles were selected for final analysis. In addition to the inclusion-exclusion criteria, the articles underwent a process of analysis and classification based on themes, methods, and results. The findings reflect a growing interest in the application of AI techniques in Legal Information Retrieval. The approaches identified focus on improving the relevance of search results and automating legal processes. There is also a progressive adoption of Natural Language Processing and machine learning techniques. Finally, a panoramic view is provided of the intersection between AI and Legal Information Retrieval. The results highlight the importance and potential of AI techniques in the legal field, while underscoring the need for deeper research and integrated approaches to address the specific challenges of Legal Information Retrieval in a technologically dynamic world.
The continuous increase of digital documents on the web creates the need to search for information patterns that allow the categorization of organizational documents to generate knowledge in an institution. An Artificial Intelligence technique for this purpose is text classification, it for its application uses labels (previously categorized documents) with supervised (with labels) or unsupervised (without labels) training models. Both traditional models with their advantages and disadvantages have been joined into semi-supervised models that extract the best qualities of each one, however, the labeling process involves resources and time that try to be optimized to improve classification accuracy. An analysis of the different semi-supervised models would show us the advantages of their training and the way how the structure of each of them affects the accuracy of their classification. In the present study, a classification structure of the semi-supervised models in the classification of documents is proposed to analyze their qualities and categorization process, through an SLR (Revision of systematic literature) that extracts performance metrics from the identified studies to perform a meta-analysis through forest plots. To define the search strategy for studies, the PICOC (Population, Intervention, Comparison, Outputs, Context) method has been used, it is supported by the research question defines a search string, which has allowed the collection of 228 research, these are filtered with the PRISMA declaration method and the determination of exclusion criteria, in this way 35 researches are selected for the present study. The analysis of the selected studies identifies a structure for the different semi-supervised learning models, and a scheme of their work process is obtained, it has been used to extract advantages, disadvantages, and performance metrics. Through a meta-analysis with forest diagrams, the classification accuracy performance of the researches in each learning model is evaluated, determining as results that regardless of the characteristics of its process, active learning (0.89) and assembled learning (0.83) present the best performance levels.