Computer Standards & Interfaces最新文献

{"title":"Grammar-obeying program synthesis: A novel approach using large language models and many-objective genetic programming","authors":"Ning Tao ,&nbsp;Anthony Ventresque ,&nbsp;Vivek Nallur ,&nbsp;Takfarinas Saber","doi":"10.1016/j.csi.2024.103938","DOIUrl":"10.1016/j.csi.2024.103938","url":null,"abstract":"<div><div>Program synthesis is an important challenge that has attracted significant research interest, especially in recent years with advancements in Large Language Models (LLMs). Although LLMs have demonstrated success in program synthesis, there remains a lack of trust in the generated code due to documented risks (e.g., code with known and risky vulnerabilities). Therefore, it is important to restrict the search space and avoid bad programs. In this work, pre-defined restricted Backus–Naur Form (BNF) grammars are utilised, which are considered ‘safe’, and the focus is on identifying the most effective technique for <em>grammar-obeying program synthesis</em>, where the generated code must be correct and conform to the predefined grammar. It is shown that while LLMs perform well in generating correct programs, they often fail to produce code that adheres to the grammar. To address this, a novel Similarity-Based Many-Objective Grammar Guided Genetic Programming (SBMaOG3P) approach is proposed, leveraging the programs generated by LLMs in two ways: (i) as seeds following a grammar mapping process and (ii) as targets for similarity measure objectives. Experiments on a well-known and widely used program synthesis dataset indicate that the proposed approach successfully improves the rate of grammar-obeying program synthesis compared to various LLMs and the state-of-the-art Grammar-Guided Genetic Programming. Additionally, the proposed approach significantly improved the solution in terms of the best fitness value of each run for 21 out of 28 problems compared to G3P.</div></div>","PeriodicalId":50635,"journal":{"name":"Computer Standards & Interfaces","volume":"92 ","pages":"Article 103938"},"PeriodicalIF":4.1,"publicationDate":"2024-11-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142663154","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"LAMB: An open-source software framework to create artificial intelligence assistants deployed and integrated into learning management systems","authors":"Marc Alier ,&nbsp;Juanan Pereira ,&nbsp;Francisco José García-Peñalvo ,&nbsp;Maria Jose Casañ ,&nbsp;Jose Cabré","doi":"10.1016/j.csi.2024.103940","DOIUrl":"10.1016/j.csi.2024.103940","url":null,"abstract":"<div><div>This paper presents LAMB (Learning Assistant Manager and Builder), an innovative open-source software framework designed to create AI-powered Learning Assistants tailored for integration into learning management systems. LAMB addresses critical gaps in existing educational AI solutions by providing a framework specifically designed for the unique requirements of the education sector. It introduces novel features, including a modular architecture for seamless integration of AI assistants into existing LMS platforms and an intuitive interface for educators to create custom AI assistants without coding skills. Unlike existing AI tools in education, LAMB provides a comprehensive framework that addresses privacy concerns, ensures alignment with institutional policies, and promotes using authoritative sources. LAMB leverages the capabilities of large language models and associated generative artificial intelligence technologies to create generative intelligent learning assistants that enhance educational experiences by providing personalized learning support based on clear directions and authoritative fonts of information. Key features of LAMB include its modular architecture, which supports prompt engineering, retrieval-augmented generation, and the creation of extensive knowledge bases from diverse educational content, including video sources. The development and deployment of LAMB were iteratively refined using a minimum viable product approach, exemplified by the learning assistant: “Macroeconomics Study Coach,” which effectively integrated lecture transcriptions and other course materials to support student inquiries. Initial validations in various educational settings demonstrate the potential that learning assistants created with LAMB have to enhance teaching methodologies, increase student engagement, and provide personalized learning experiences. The system's usability, scalability, security, and interoperability with existing LMS platforms make it a robust solution for integrating artificial intelligence into educational environments. LAMB's open-source nature encourages collaboration and innovation among educators, researchers, and developers, fostering a community dedicated to advancing the role of artificial intelligence in education. This paper outlines the system architecture, implementation details, use cases, and the significant benefits and challenges encountered, offering valuable insights for future developments in artificial intelligence assistants for any sector.</div></div>","PeriodicalId":50635,"journal":{"name":"Computer Standards & Interfaces","volume":"92 ","pages":"Article 103940"},"PeriodicalIF":4.1,"publicationDate":"2024-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142663153","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Integration of fuzzy-weighted zero-inconsistency and fuzzy decision by opinion score methods under a q-rung orthopair environment: A distribution case study of COVID-19 vaccine doses","authors":"A.S. Albahri ,&nbsp;O.S. Albahri ,&nbsp;A.A. Zaidan ,&nbsp;Alhamzah Alnoor ,&nbsp;H.A. Alsattar ,&nbsp;Rawia Mohammed ,&nbsp;A.H. Alamoodi ,&nbsp;B.B. Zaidan ,&nbsp;Uwe Aickelin ,&nbsp;Mamoun Alazab ,&nbsp;Salem Garfan ,&nbsp;Ibraheem Y.Y. Ahmaro ,&nbsp;M.A. Ahmed","doi":"10.1016/j.csi.2021.103572","DOIUrl":"10.1016/j.csi.2021.103572","url":null,"abstract":"<div><p>Owing to the limitations of Pythagorean fuzzy and intuitionistic fuzzy sets, scientists have developed a distinct and successive fuzzy set called the q-rung orthopair fuzzy set (q-ROFS), which eliminates restrictions encountered by decision-makers in multicriteria decision making (MCDM) methods and facilitates the representation of complex uncertain information in real-world circumstances. Given its advantages and flexibility, this study has extended two considerable MCDM methods the fuzzy-weighted zero-inconsistency (FWZIC) method and fuzzy decision by opinion score method (FDOSM) under the fuzzy environment of q-ROFS. The extensions were called q-rung orthopair fuzzy-weighted zero-inconsistency (q-ROFWZIC) method and q-rung orthopair fuzzy decision by opinion score method (q-ROFDOSM). The methodology formulated had two phases. The first phase ‘development’ presented the sequential steps of each method thoroughly.The q-ROFWZIC method was formulated and used in determining the weights of evaluation criteria and then integrated into the q-ROFDOSM for the prioritisation of alternatives on the basis of the weighted criteria. In the second phase, a case study regarding the MCDM problem of coronavirus disease 2019 (COVID-19) vaccine distribution was performed. The purpose was to provide fair allocation of COVID-19 vaccine doses. A decision matrix based on an intersection of ‘recipients list’ and ‘COVID-19 distribution criteria’ was adopted. The proposed methods were evaluated according to systematic ranking assessment and sensitivity analysis, which revealed that the ranking was subject to a systematic ranking that is supported by high correlation results over different scenarios with variations in the weights of criteria.</p></div>","PeriodicalId":50635,"journal":{"name":"Computer Standards & Interfaces","volume":"80 ","pages":"Article 103572"},"PeriodicalIF":5.0,"publicationDate":"2022-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1016/j.csi.2021.103572","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"10390432","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}