Niyati Chhaya

Identifying Suggestions for Improvement of Product Features from Online Product Reviews

Online forums are used to share experiences and opinions about products and services. These forums range from review sites such as Amazon (www.amazon.com) to online social networks such as Twitter (www.twitter.com). The user-generated content in these platforms capture the users’ opinions and sentiments. In this work, we explore the problem of identifying suggestions from text content. The paper first defines suggestive intent and then presents a supervised learning approach to identify text that contains suggestive intent. The results show high accuracy with a F1 score of 0.93.

Probabilistic Deduplication of Anonymous Web Traffic

Cookies and log in-based authentication often provide incomplete data for stitching website visitors across multiple sources, necessitating probabilistic deduplication. We address this challenge by formulating the problem as a binary classification task for pairs of anonymous visitors. We compute visitor proximity vectors by converting categorical variables like IP addresses, product search keywords and URLs with very high cardinalities to continuous numeric variables using the Jaccard coefficient for each attribute. Our method achieves about 90% AUC and F-scores in identifying whether two cookies map to the same visitor, while providing insights on the relative importance of available features in Web analytics towards the deduplication process.

BATframe: An Unsupervised Approach for Domain-Sensitive Affect Detection

Generic sentiment and emotion lexica are widely used for the fine–grained analysis of human affect from text. In order to accurately detect affect, there is a need for domain intelligence, that enables understanding of the perceived interpretation of the same words in varied contexts. Recent work has focused on automatically inducing the polarity of given terms in changing contexts. We propose an unsupervised approach for the construction of domain–specific affect lexica along these lines. The algorithm is seeded with existing standard lexica and expanded based on context–relevant word associations. Experiments show that our lexicon provides better coverage than standard lexica on both short as well as long texts, and corresponds well with human–annotated affect values. Our framework outperforms the state–of–the–art generic and domain–specific approaches with a precision of over 70% for the emotion detection task on the SemEval 2007 Affect Corpus.

Leveraging Site Search Logs to Identify Missing Content on Enterprise Webpages

Online visitors often do not find the content they were expecting on specific pages of a large enterprise website, and subsequently search for it in site’s search box. In this paper, we propose methods to leverage website search logs to identify missing or expected content on webpages on the enterprise website, while showing how several scenarios make this a non-trivial problem. We further discuss how our methods can be easily extended to address concerns arising from the identified missing content.

EnTwine: Feature Analysis and Candidate Selection for Social User Identity Aggregation

Organizations measure their social audience based on the number of users, fans, and followers on social media. Every social media platform has its user identity and a single user is present across varied platforms. Due to the disconnected user profiles, identifying duplicate users across media is non-trivial. There is a need to create a complete view of a user for various applications such as targeting and user profile construction. This view is not easily available due to the individual identities. In this work, we explore the feature space across social media that can be leveraged for intelligent user identity aggregation. Further, we present a two-phased unified identity creation process using our feature analysis, unsupervised candidate selection, and supervised user matching algorithms on four different social networks.