Search Engine Optimization by Re-Ranking the Product Search Result Based on User Click Data

Blibli.com provides a search engine for its customers. It used Solr search engine with only plain BM25 similarity function which is based on probability. In order to improve search engine performance, this research tried to implement an algorithm that involves user click data to re-rank the search results of Solr search engine. We used 1386 product data given by Blibli.com and worked with 5 judges to label the data. The implementation of clickthrough data algorithm which was built into a plugin in Solr search engine gave a promising result, that there is an increased of the mean average precision as much as 21%. Hence, we concluded that the clickthrough data algorithm has increased the rank quality of Blibli.com product search result in Solr search engine. However, there are still many irrelevant products retrieved in the Solr search engine, although its ranking position were below the relevant products.