Search

Location information is very important in the real world. How to detect users' locations and further movement trajectories automatically is significant for many location-based services such as dietary recommendation and tourism planning. With the rapid development of social media such as Sina Weibo and Twitter, more and more people publish posts at any time which contain their real-time location information. This makes it possible to detect users' movement trajectories automatically by social media. In this paper, we propose a novel method to detect city-level locations and further identify movement trajectory for a given user in social media. Our efforts are made in two phases: (1) location detection, and (2) movement trajectory identification. Phase (1) combines two components, i.e. a Bayes model based on words distribution over locations and the social relationships of users in social media, to detect or infer users' locations. The former one is utilized to judge whether the content of a post contains explicit/implicit location information and further to detect the location. The latter one assumes that friends tend to gather together when they talk about the same thing. Thus, it infers a user's locations from his friends' detected locations in social media. Phase (2) constructs the user's movement trajectory by smoothing these detected locations in first phase according to both semantic context of posts and transfer time between locations. Experiments on real dataset from Sina Weibo demonstrate that our approach can outperform baseline methods significantly in terms of Precision, Recall, F1, Error Distance ( ErrDist), Detect Ratio ( DeteRatio), T-Precision, T-Recall and T-F1, respectively. [ABSTRACT FROM AUTHOR]

How to understand intents behind user queries is crucial towards improving the performance of Web search systems. NTCIR-11 IMine task focuses on this problem. In this paper, we address the NTCIR-11 IMine task with two phases referred to as Query Intent Mining ( QIM) and Query Intent Ranking ( QIR). (I) QIM is intended to mine users' potential intents by clustering short text fragments related to the given query. (II) QIR focuses on ranking those mined intents in a proper way. Two challenges exist in handling these tasks. (II) How to precisely estimate the intent similarity between user queries which only consist of a few words. (2) How to properly rank intents in terms of multiple factors, e.g. relevance, diversity, intent drift and so on. For the first challenge, we first investigate two interesting phenomena by analyzing query logs and document datasets, namely ' Same-Intent-Co-Click' ( SICC) and ' Same-Intent-Similar-Rank' ( SISR). SICC means that when users issue different queries, these queries represent the same intent if they click on the same URL. SISR means that if two queries denote the same intent, we should get similar search results when issuing them to a search engine. Then, we propose similarity functions for QIM based on the two phenomena. For the second challenge, we propose a novel intent ranking model which considers multiple factors as a whole. We perform extensive experiments and an interesting case study on the Chinese dataset of NTCIR-11 IMine task. Experimental results demonstrate the effectiveness of our proposed approaches in terms of both QIM and QIR. [ABSTRACT FROM AUTHOR]

Recommender systems with social networks have been well studied in recent years. However, most of these methods ignore the social contextual information among users and items, which is significant and useful for predicting users' preferences in many recommendation problems. Moreover, most existing social recommendation methods have been proposed for the scenarios where users can provide explicit ratings. But in fact, the explicit feedback is not always available, most of the feedback in real social networks is not explicit but implicit. Motivated by above observations, we propose a unified ranking framework fusing social contextual information and common social relations for implicit feedback. Specifically, we first extend the user latent features by the implicit interest deduced from social context, and then we integrate the common social relations as factorization terms to further improve recommendation quality. Finally, we optimize our model in a Bayesian personalized ranking framework. The experiments on real-world dataset show that our approach outperforms the other state-of-the-art algorithms in terms of AUC, NDCG and Pre@3. This result demonstrates the importance of social context and common social relations for the formation of the implicit ratings. [ABSTRACT FROM AUTHOR]

During a two-day strategic workshop in February 2018, 22 information retrieval researchers met to discuss the future challenges and opportunities within the field. The outcome is a list of potential research directions, project ideas, and challenges. This report describes the major conclusions we have obtained during the workshop. A key result is that we need to open our mind to embrace a broader IR field by rethink the definition of information, retrieval, user, system, and evaluation of IR. By providing detailed discussions on these topics, this report is expected to inspire our IR researchers in both academia and industry, and help the future growth of the IR research community. [ABSTRACT FROM AUTHOR]