Multi-view Clustering Trees for Search and Retrieval in Customer Product Forums

Customers of large enterprises post threads in companysupported online forums to discuss solutions to product malfunctions, errors and problems. The ability to retrieve the most relevant threads in response to a customer’s search query in the product forums requires robust search capabilities. Unfortunately, the lack of (recommendation) links between the threads in the product discussion forums makes it infeasible to use web search algorithms such as PageRank in these forums. We provide a keyword extraction technique based on term co-occurrences that performs better than traditional tfidfbased techniques. We further provide a search and retrieval algorithm to retrieve the relevant threads in response to a search query in product discussion forums. Our approach is based on a hierarchical, multi-view (thread title and thread content) clustering of the threads. We evaluate our approach and compare with existing statistical algorithms on threads selected from HP’s customer support forum.