Searching Behavior Analysis of Online Shopping Based on Information Content of Query Words

With the spread of the internet as social infrastructure, more and more people are shopping online. Online sites that formerly dealt with such specific products as books and clothing have also expanded to mall-type shopping sites by incorporating various kinds of stores. As a result, searching for products has become more complicated and prolonged. In this paper, we propose a method that models product-searching behavior based on the transition of the search words input by users. Since a query is generally composed of one or more search words, their information content is calculated in advance from query logs. Thus, varying the information content of the user's query sequences can be classified as a model of user searching behaviors. From analysis results using actual data, we confirmed that our proposed method effectively models product-searching behavior.