Task-based clustering on search queries

Task extraction on query logs is one of the important and interesting topics used on search engines and many search-based applications. Similar queries entered by users can be aggregated according to their various features to make meaningful tasks. Task extraction is important for providing suggestions in the direction of the user's intention, in the search text completion, in returning the correct results for the domain being searched. Session information, clicked document contents and query entities are used for feature extraction in existing approaches. In recent studies, task extraction is made by the Wikipedia category hierarchies that is used at clustering level. In this study, entity categories are used for feature extraction instead of clustering phase. Therefore, it is aimed to measure sentimental similarity between search queries on categories precisely. In addition, the queries are clustered by central based and density based clustering methods with the different combination of feature sets. The evaluation of the methods are obtained by considering the similarities within cluster members and between clusters' centers. As a result of this work, entity and category vectors generated by word2vec method are treated as query feature and task based clustering performance is increased.