Hossein Fani

Semantics-Enabled User Interest Detection from Twitter

Social networks enable users to freely communicate with each other and share their recent news, ongoing activities or views about different topics. As a result, user interest detection from social networks has been the subject of increasing attention. Some recent works have proposed to enrich social posts by annotating them with unambiguous relevant ontological concepts extracted from external knowledge bases and model user interests as a bag of concepts. However, in the bag of concepts approach, each topic of interest is represented as an individual concept that is already predefined in the knowledge base. Therefore, it is not possible to infer fine-grained topics of interest, which are only expressible through a collection of multiple concepts or emerging topics, which are not yet defined in the knowledge base. To address these issues, we view each topic of interest as a conjunction of several concepts, which are temporally correlated on Twitter. Based on this, we extract active topics within a given time interval and determine a users inclination towards these active topics. We demonstrate the effectiveness of our approach in the context of a personalized news recommendation system. We show through extensive experimentation that our work is able to improve the state of the art.

Inferring Implicit Topical Interests on Twitter

Inferring user interests from their activities in the social network space has been an emerging research topic in the recent years. While much work is done towards detecting explicit interests of the users from their social posts, less work is dedicated to identifying implicit interests, which are also very important for building an accurate user model. In this paper, a graph based link prediction schema is proposed to infer implicit interests of the users towards emerging topics on Twitter. The underlying graph of our proposed work uses three types of information: user’s followerships, user’s explicit interests towards the topics, and the relatedness of the topics. To investigate the impact of each type of information on the accuracy of inferring user implicit interests, different variants of the underlying representation model are investigated along with several link prediction strategies in order to infer implicit interests. Our experimental results demonstrate that using topics relatedness information, especially when determined through semantic similarity measures, has considerable impact on improving the accuracy of user implicit interest prediction, compared to when followership information is only used.

Finding Diachronic Like‐Minded Users

User communities in social networks are usually identified by considering explicit structural social connections between users. While such communities can reveal important information about their members such as family or friendship ties and geographical proximity, just to name a few, they do not necessarily succeed at pulling like‐minded users that share the same interests together. Therefore, researchers have explored the topical similarity of social content to build like‐minded communities of users. In this article, following the topic‐based approaches, we are interested in identifying communities of users that share similar topical interests with similar temporal behavior. More specifically, we tackle the problem of identifying temporal (diachronic) topic‐based communities, i.e., communities of users who have a similar temporal inclination toward emerging topics. To do so, we utilize multivariate time series analysis to model the contributions of each user toward emerging topics. Further, our modeling is completely agnostic to the underlying topic detection method. We extract topics of interest by employing seminal topic detection methods; one graph‐based and two latent Dirichlet allocation‐based methods. Through our experiments on Twitter data, we demonstrate the effectiveness of our proposed temporal topic‐based community detection method in the context of news recommendation, user prediction, and document timestamp prediction applications, compared with the nontemporal as well as the state‐of‐the‐art temporal approaches.

Community detection in social networks

Online social networks have become a fundamental part of the global online experience. They facilitate different modes of communication and social interactions, enabling individuals to play social roles that they regularly undertake in real social settings. In spite of the heterogeneity of the users and interactions, these networks exhibit common properties. For instance, individuals tend to associate with others who share similar interests, a tendency often known as homophily, leading to the formation of communities. This entry aims to provide an overview of the definitions for an online community and review different community detection methods in social networks. Finding communities are beneficial since they provide summarization of network structure, highlighting the main properties of the network. Moreover, it has applications in sociology, biology, marketing and computer science which help scientists identify and extract actionable insight.

Predicting Users’ Future Interests on Twitter

In this paper, we address the problem of predicting future interests of users with regards to a set of unobserved topics in microblogging services which enables forward planning based on potential future interests. Existing works in the literature that operate based on a known interest space cannot be directly applied to solve this problem. Such methods require at least a minimum user interaction with the topic to perform prediction. To tackle this problem, we integrate the semantic information derived from the Wikipedia category structure and the temporal evolution of user’s interests into our prediction model. More specifically, to capture the temporal behaviour of the topics and user’s interests, we consider discrete intervals and build user’s topic profile in each time interval separately. Then, we generalize users’ interests that have been observed over several time intervals by transferring them over the Wikipedia category structure. Our approach not only allows us to generalize users’ interests but also enables us to transfer users’ interests across different time intervals that do not necessarily have the same set of topics. Our experiments illustrate the superiority of our model compared to the state of the art.

Lexical Semantic Relatedness for Twitter Analytics

Existing work in the semantic relatedness literature has already considered various information sources such as WordNet, Wikipedia and Web search engines to identify the semantic relatedness between two words. We will show that existing semantic relatedness measures might not be directly applicable to microblogging content such as tweets due to i) the informality and short length of microblogging content, which can lead to shift in the meaning of words when used in microblog posts, ii) the presence of non-dictionary words that have their semantics defined/evolved by the Twitter community. Therefore, we propose the Twitter Space Semantic Relatedness (TSSR) technique that relies on the latent relation hypothesis to measure semantic relatedness of words on Twitter. We construct a graph representation of terms in tweets and apply a random walk procedure to produce a stationary distribution for each word, which is the basis for relatedness calculation. Our experiments examine TSSR from three different perspectives and show that TSSR is better suited for Twitter analytics compared to the standard semantic relatedness techniques.

Entity linking of tweets based on dominant entity candidates

Entity linking, also known as semantic annotation, of textual content has received increasing attention. Recent works in this area have focused on entity linking on text with special characteristics such as search queries and tweets. The semantic annotation of tweets is specially proven to be challenging given the informal nature of the writing and the short length of the text. In this paper, we propose a method to perform entity linking on tweets built based on one primary hypothesis. We hypothesize that while there are formally many possible entity candidates for an ambiguous mention in a tweet, as listed on the disambiguation page of the corresponding entity on Wikipedia, there are only few entity candidates that are likely to be employed in the context of Twitter. Based on this hypothesis, we propose a method to identify such dominant entity candidates for each ambiguous mention and use them in the annotation process. Particularly, our proposed work integrates two phases (i) dominant entity candidate detection, which applies community detection methods for finding the dominant candidates of ambiguous mentions; and (ii) named entity disambiguation that links a tweet to entities in Wikipedia by only considering the identified dominant entity candidates. Our investigations show that: (1) there are only very few entity candidates for each ambiguous mention in a tweet that need to be considered when performing disambiguation. This helps us limit the candidate search space and hence noticeably reduce the entity linking time; (2) limiting the search space to only a subset of disambiguation options will not only improve entity linking execution time but will also lead to improved accuracy of the entity linking process when the main entity candidates of each mention are mined from a temporally aligned corpus. We show that our proposed method offers competitive results with the state-of-the-art methods in terms of precision and recall on widely used gold standard datasets while significantly reducing the time for processing each tweet.

Stopword Detection for Streaming Content

The removal of stopwords is an important preprocessing step in many natural language processing tasks, which can lead to enhanced performance and execution time. Many existing methods either rely on a predefined list of stopwords or compute word significance based on metrics such as tf-idf. The objective of our work in this paper is to identify stopwords, in an unsupervised way, for streaming textual corpora such as Twitter, which have a temporal nature. We propose to consider and model the dynamics of a word within the streaming corpus to identify the ones that are less likely to be informative or discriminative. Our work is based on the discrete wavelet transform (DWT) of word signals in order to extract two features, namely scale and energy. We show that our proposed approach is effective in identifying stopwords and improves the quality of topics in the task of topic detection.

An Ontology for Describing Security Events.

Mining security events helps with better precautionary planning for community safety. However, incident records are expressed in diverse and application de p ndent formats which impedes common comprehension for auto matic knowledge extraction and reasoning. In this paper, we present Security Incident Ontology, SIO, a novel light-weight domain ontology for security incidents. We use Timeline to annotate the temporal facts of incidents and adopt Event to repr esent any security issues from indecent behavior to assault t o more adverse crime which raise the security alarm in a community. It will present a unique way to the security incident detec tors, a police officer, Robocops, or intelligent CCTV cameras, to report security events. We use SIO in populating security incident notifications of Integrated Risk Management (IRM) at Ryerson University to evaluate its competency, for Ryerson University campus has both business and housing area in the vi cinity and encompass not only high rate, but also wide variety of different security issues. SIO is developed in OWL 2 with Pro tégé. Ontology; Semantic Web; OWL; Security Incident; Event.