Using machine learning algorithms in Microsoft Azure ML to improve system search
Jonida Shehu1*, Silvana Greca2, Endri Xhina3
1,2,3 University of Tirana, Faculty of Natural Sciences, Informatics Department
Machine learning algorithms have revolutionized predictive analysis, natural language processing, image classification, and information retrieval. Semantic AI is a new concept of using the power of machine learning and knowledge graphs for better search and system recommendations. This paper combines the knowledge graph information from an ontology with ML algorithms in a system where citizens can search for e-government services. The ontology describes linked entities of government institutions, legislations, public services, documentation, and their attributes. Service ranking rating and previous searches saved on the ontology provide users with better experience and search results. The experiment we model in Microsoft Azure ML uses data from the ontology and classifies public services with more accuracy and precision using machine learning algorithms by assigning weights to essential words when searching.
Keywords: ML algorithms; ontology; semantic search; e-government;
Machine learning addresses how to build computers that improve automatically through experience. It is one of today’s most rapidly growing technical fields, lying at the intersection of computer science and statistics and the core of artificial intelligence and data science (Kravets et al., 2015). Machine learning algorithms have been applied in different fields such as banking, e-government, medical, physics, and e-learning.
E-Government refers to the use of internet technology as a platform for exchanging information, providing services, and transacting with citizens, businesses, and other arms of government (Kamal, 2009). As administrative agencies serving the people, government agencies provide people with convenient and fast services. With the popularization and development of information technology, traditional inefficient government work methods can no longer adapt to the development of social information and cannot satisfy people’s convenience and the need for fast administrative services;(Zhao, 2021). Every citizen can have a personalized experience with the use of ML applications.
Although there is a huge number of research in the literature related to ML applications, there is a lack of a comprehensive study focusing on the usage of this technology within governmental applications. (Charalampos et.al 2019).
Volume 6.No.1(2022): April – (Humanities Session)
ISSN 2661-2666 (Online) International Scientific Journal Monte (ISJM)
ISSN 2661-264X (Print)
DOI : 10.33807/monte.20222486
This is an open access article under the CC BY-NC-ND license (creativecommons.org/licenses/by-nc-nd/4.0/)