A probabilistic mixture model for mining and analyzing product search log

Abstract

The booming of e-commerce in recent years has led to the generation of large amounts of product search log data. Product search log is a unique new data with much valuable information and knowledge about user preferences over product attributes that is often hard to obtain from other sources. While regular search logs (e.g., Web search logs) contain click-throughs for unstructured text documents (e.g., web pages), product search logs contain clickth-roughs for structured entities defined by a set of attributes and their values. For instance, a laptop can be defined by its size, color, cpu, ram, etc. Such structures in product entities offer us opportunities to mine and discover detailed useful knowledge about user preferences at the attribute level, but they also raise significant challenges for mining due to the lack of attribute-level observations.In this paper, we propose a novel probabilistic mixture model for attribute-level analysis of product search logs. The model is based on a generative process where queries are generated by a mixture of unigram language models defined by each attribute-value pair of a clicked entity. The model can be efficiently estimated using the ExpectationMaximization (EM) algorithm. The estimated parameters, including the attribute-value language models and attributevalue preference models, can be directly used to improve product search accuracy, or aggregated to reveal knowledge for understanding user intent and supporting business intelligence. Evaluation of the proposed model on a commercial product search log shows that the model is effective for mining and analyzing product search logs to discover various kinds of useful knowledge.

Conversational Product Search Based on Negative Feedback

“…However, free-form user queries are difficult to structure. To support search based on keyword queries, Duan et al [8,9] extended the query likelihood [20] method by assuming that queries are generated from a mixture of two language models, one of the background corpus, the other of products conditioned on their specifications. This approach still cannot solve the vocabulary mismatch problem between user queries and product descriptions or reviews.…”

A Zero Attention Model for Personalized Product Search

“…Product search is an important problem that has been widely studied in the research communities of Data Mining and Information Retrieval. Early studies mainly focus on how to construct effective retrieval systems that support search for structured product information [10][11][12]. For example, Lim et al [23] propose to conduct product retrieval with a facet search engine built on structured representations of products in relational databases such as brands, prices, categories, etc.…”

A Study of Context Dependencies in Multi-page Product Search

“…Facets of products have been used for product search [9,14]. Language model based approaches have been studied to support keyword search [3]. Later, to further solve vocabulary mismatch, models that measure semantic match between queries and products based on reviews have been proposed [1,13].…”

“…Similar to [2,12], we represent the relevant set with the centroid of each item in the set and average item embeddings are used to represent the centroid, denoted as E(C 1:t ). 3 Overall Context Embeddings. We use a convex combination of user, query, and click embeddings as the representation of overall context E(S t ).…”

“…The attention weights then act as the basis for re-ranking. They can be computed as: 3 We also tried an attention mechanism to weight each clicked item according to the query and represent the user preference with a weighted combination of clicked items. However, this method is not better than combining clicks with equal weights.…”