Online review sites and opinion forums contain a wealth of information regarding user preferences and experiences over multiple product domains. This information can be leveraged to obtain valuable insights using data mining approaches such as sentiment analysis. In this work we examine online user reviews within the pharmaceutical field. Online user reviews in this domain contain information related to multiple aspects such as effectiveness of drugs and side effects, which make automatic analysis very interesting but also challenging. However, analyzing sentiments concerning the various aspects of drug reviews can provide valuable insights, help with decision making and improve monitoring public health by revealing collective experience. In this preliminary work we perform multiple tasks over drug reviews with data obtained by crawling online pharmaceutical review sites. We first perform sentiment analysis to predict the sentiments concerning overall satisfaction, side effects and effectiveness of user reviews on specific drugs. To meet the challenge of lacking annotated data we further investigate the transferability of trained classification models among domains, i.e. conditions, and data sources. In this work we show that transfer learning approaches can be used to exploit similarities across domains and is a promising approach for cross-domain sentiment analysis.

With the rapid adoption of online shopping, academic research in the eCommerce domain has gained traction. However, significant research challenges remain, spanning from classic eCommerce search problems such as matching textual queries to multi-modal documents and ranking optimization for two-sided marketplaces to human-computer interaction and recommender systems for discovery and browsing. These research areas are important for understanding customer behavior, driving engagement, and improving product discoverability and conversion. In this article we identify the challenges and highlight research opportunities to improve the eCommerce customer experience.

In search and recommendation, diversifying the multi-aspect search results could help with reducing redundancy, and promoting results that might not be shown otherwise. Many previous methods have been proposed for this task. However, previous methods do not explicitly consider the uniformity of the number of the items' classes, or evenness, which could degrade the search and recommendation quality. To address this problem, we introduce a novel method by adapting the Simpson's Diversity Index from biology, which enables a more effective and efficient quadratic search result diversification algorithm. We also extend the method to balance the diversity between multiple aspects through weighted factors and further improve computational complexity by developing a fast approximation algorithm. We demonstrate the feasibility of the proposed method using the openly available Kaggle shoes competition dataset. Our experimental results show that our approach outperforms previous state of the art diversification methods, while reducing computational complexity.

You will be contacted as soon as possible.

No abstract