Supporting keyword search in product database

Abstract

The ability to let users search for products conveniently in product database is critical to the success of e-commerce. Although structured query languages (e.g. SQL) can be used to effectively access the product database, it is very difficult for end users to learn and use. In this paper, we study how to optimize search over structured product entities (represented by specifications) with keyword queries such as "cheap gaming laptop". One major difficulty in this problem is the vocabulary gap between the specifications of products in the database and the keywords people use in search queries. To solve the problem, we propose a novel probabilistic entity retrieval model based on query generation, where the entities would be ranked for a given keyword query based on the likelihood that a user who likes an entity would pose the query. Different ways to estimate the model parameters would lead to different variants of ranking functions. We start with simple estimates based on the specifications of entities, and then leverage user reviews and product search logs to improve the estimation. Multiple estimation algorithms are developed based on Maximum Likelihood and Maximum a Posteriori estimators. We evaluate the proposed product entity retrieval models on two newly created product search test collections. The results show that the proposed model significantly outperforms the existing retrieval models, benefiting from the modeling of attribute-level relevance. Despite the focus on product retrieval, the proposed modeling method is general and opens up many new opportunities in analyzing structured entity data with unstructured text data. We show the proposed probabilistic model can be easily adapted for many interesting applications including facet generation and review annotation.

A Product Feature-Based User-Centric Ranking Model for E-Commerce Search

“…For instance, the query "cheap PC gamer" might be difficult to solve by only comparing the query text with the product description since it requires reasoning over the search intent towards a particular product feature, namely the price. To tackle this challenge, Duan et al [9] propose to represent both products and users through an entity-based representation in which each entity is formalized as a pair of key-value. The product retrieval is then performed through a probabilistic model which estimates the relevance at the level of attribute preferences.…”

“…In contrast to [9,10,16] which focused on the interest of a particular user, our proposed user engagement metric leverages from the crowd.…”

“…On one hand, the proposed approaches focus mainly on the product fields, namely its category and its description [4,18]. On the other hand, users' preferences and search intent are emphasized leading to a user-centered search process [9].…”

“…The latter may be characterized by several variations, accessories, and/or complementary products. Combining both approaches (product features [4,18] and user-centered [9]), our model ranks products with respect to their descriptive fields and category as well as their popularity highlighting the user engagement toward the product [14]. We evaluate our model while participating to the product search track of the Living Labs for IR (LL4IR) [17] of CLEF 2015 [11] and present also some analysis aiming at understanding the user engagement factor.…”

Supporting Keyword Search for Image Retrieval with Integration of Probabilistic Annotation

“…The emission probabilities are computed for each keyword and for each database attribute by applying the search function over full text indexes provided by the DBMS. In [11], we studied how to optimize search of structured product entities (represented by specifications) with keyword queries such as "cheap gaming laptop". They propose a novel probabilistic entity retrieval model based on query generation, where the entities would be ranked for a given keyword query based on the likelihood that a user who likes an entity would pose the query.…”