Comparative study for better result on query suggestion of article searching with MySQL pattern matching and Jaccard similarity

Article searching is the process to find the appropriate article based on user description or keywords. For the traditional article searching, user types keywords then press the button and the result presents, but unfortunately the process needs more time and the search result is not accurate based on user keywords. Query suggestion makes it easy to search the data without submitting the data in a traditional way. There are several methods that can be used in query suggestion thus it is necessary to know which method should be used. This paper describes the comparison between MySQL pattern matching and Jaccard similarity for query suggestion with the parameters, they are processing time, the proximity of keywords with the suggestion, the proximity ratings and sorted from nearest data, the strength of data retrieval based on the number of words entered, and the strength of data retrieval from a number of data that can be suggested. MySQL pattern matching and Jaccard similarity can be used as query suggestion with the processing time is 2.5 milliseconds for MySQL pattern matching and 7 milliseconds for Jaccard similarity with 492 articles. Query suggestion using MySQL pattern matching takes an average of 1.8 milliseconds for 100 articles, 2 milliseconds for 200 articles, 2.175 milliseconds for 300 articles, while Jaccard similarity 3.175 milliseconds for 100 articles, 4.025 milliseconds for 200 articles, 5.05 milliseconds for 300 articles. Proximity ratings for MySQL Pattern Matching cannot be calculated since there is no counting process, but in Jaccard similarity the system displays 0.8 for jaccard index and display 1 for jaccard index if the keywords are 100% similar to the data on database. The result of this study is Jaccard similarity query suggestion produces more accurate search result but has a longer processing time than MySQL pattern matching.