Affiliations: Department of Information, Business College of Shanxi University, Taiyuan, China
Correspondence:
[*]
Corresponding author. Xiaohui Ma, Department of Information, Business College of Shanxi University, 030031, Taiyuan, China. E-mail: [email protected].
Abstract: Aggregated search is the task of integrating results from potentially multiple specialized search services, or verticals (images, videos, news, etc.), into the Web search results. Major search engines perform what is known as Aggregated Search. Aggregated search is relatively new and its advantages need to be evaluated. With the increasing size of the data set on the network, different results retrieved from different dimensions are sometimes entirely different for a given search problem. In this paper, a new aggregated search algorithm was proposed. Firstly, the retrieval data in the vertical domain was textualized, and Doc2Vec makes the vector representation of the data. Then the results are aggregated and output by dimension reduction and density clustering. The experimental results show that the model achieves good accuracy in 20 given queries and significantly improves the aggregated search results. Discussion about the results also allowed us to identify some useful thoughts concerning the evaluation of AS approaches.
Keywords: Aggregated search, dimension reduction, density clustering, information extraction
