Metha Herliani
Teknik Informatika STMIK ATMA LUHUR PANGKALPINANG
Jl. Jenderal Sudirman Selindung Lama Pangkalpinang Kepulauan Babel
Email: metha_herliani@yahoo.com
ABSTRACT
As the development of technology, the impact on libraries is that the need for information must also be fast and can allow a user without having to type the exact title of the book you are looking for. Search engine application that the author made the book aims to facilitate users in terms of finding relevant book titles or irrelevant so the user will be able to see the degree of similarity of the titles sought from the results. The method I use in this application is a method of vector space model which in this method the document search results will be sorted based on the weight of the search words contained in the document. One weighting algorithm is an algorithm tf – idf influenced by the frequency of occurrence of the word in a document and the frequency of documents that have the word and text pemfrofilan trigram that every character in the query inputted in pieces in the constitution. To determine the relevant level of an inputted document matches the category of the similarity of a document by using Jaccard algorithm. At this stage of measuring the effectiveness of obtaining the information required testing accuracy with precision and testing formulas acquisition by formula recall. Can be inferred using the method the user more easily find the desired book titles without the need to type in a search query exactly the same titles.
Keywords: library, trigram, vector space model, weighting text, recall, precision, query
