Skip to main content
main-content

Über dieses Buch

This edited volume offers a clear in-depth overview of research covering a variety of issues in social search and recommendation systems. Within the broader context of social network analysis it focuses on important and up-coming topics such as real-time event data collection, frequent-sharing pattern mining, improvement of computer-mediated communication, social tagging information, search system personalization, new detection mechanisms for the identification of online user groups, and many more. The twelve contributed chapters are extended versions of conference papers as well as completely new invited chapters in the field of social search and recommendation systems. This first-of-its kind survey of current methods will be of interest to researchers from both academia and industry working in the field of social networks.

Inhaltsverzeichnis

Frontmatter

Adaptive Identification of Hashtags for Real-Time Event Data Collection

Abstract
The widespread use of microblogging services, such as Twitter, makes them a valuable tool to correlate people’s personal opinions about popular public events. Researchers have capitalized on such tools to detect and monitor real-world events based on this public, social, perspective. Most Twitter event analysis approaches rely on event tweets collected through a set of predefined keywords. In this paper, we show that the existing data collection approaches risk losing a significant amount of event-relevant information. We propose a refined adaptive crawling model, to detect emerging popular topics, using hashtags, and monitor them to retrieve greater amounts of highly associated data for the events of interest. The proposed adaptive crawling model expands the queries periodically by analyzing the traffic pattern of hashtags collected from a live Twitter stream. We evaluated this adaptive crawling model with a real-world event. Based on the theoretical analysis, we tuned the parameters and ran three crawlers, including one baseline and two adaptive crawlers, during the 2013 Glastonbury music festival. Our analysis shows that adaptive crawling based on a Refined Keyword Adaptation algorithm outperforms the others. It collects the most comprehensive set of keywords, and with the minimal introduction of noise.
Xinyue Wang, Laurissa Tokarchuk, Felix Cuadrado, Stefan Poslad

Comparison of Emoticon Recommendation Methods to Improve Computer-Mediated Communication

Abstract
This paper describes the development of an emoticon recommendation system based on users’ emotional statements. In order to develop this system, an innovative emoticon database consisting of a table of emoticons with points expressed from each of 10 distinctive emotions was created. An evaluation experiment showed that our proposed system achieved an improvement of 28.1 points over a baseline system, which recommends emoticons based on users’ past emoticon selection. We also integrated the proposed and baseline systems, leading to a performance improvement of approximately 73.0 % in the same experiment. Evaluation of respondents’ perceptions of the three systems utilizing an SD scale and factor analysis is also described in this paper.
Yuki Urabe, Rafal Rzepka, Kenji Araki

Accuracy Versus Novelty and Diversity in Recommender Systems: A Nonuniform Random Walk Approach

Abstract
In this chapter, we focus on recommender systems that are enhanced with social information in the form of trust statements between their users. The trust information may be processed in a number of ways, including the random walks in the social graph, where every step in the walk is chosen almost uniformly at random from the available choices. Although this strategy yields satisfactory results in terms of the novelty and the diversity of the produced recommendations, it exhibits poor accuracy because it does not fully exploit the similarity information among users and items. Our work tries to model user-to-user and user-to-item relation as a probability distribution using a novel approach based on Rejection Sampling in order to decide its next step (biased random walk). Some initial results on reference datasets indicate that a satisfying trade-off among accuracy, novelty, and diversity is achieved.
Georgios Alexandridis, Georgios Siolas, Andreas Stafylopatis

Social Network Derived Credibility

Abstract
The increasing use of social media results in users that must ascertain the truthfulness of information that they encounter from unknown sources using a variety of indicators (e.g., explicit ratings, profile information, etc.). Through human-subject experimentation with an online social network-style platform, we focus on the determination of credibility in ego-centric networks, where participants are able to observe salient social network properties, such as degree centrality and geodesic distance. Using manipulated social network graphs, we find that corroboration and degree centrality are most utilized by subjects as indicators of credibility. utilized by subjects as indicators of credibility. We discuss the implications of the use of social network structural properties, use principal components analysis to visualize the reduced dimensional feature space, and analyze how credibility changes per property according to the “Big 5” theory of personality.
Erica J. Briscoe, Darren Scott Appling, Heather Hayes

Anonymizing Social Network Data for Maximal Frequent-Sharing Pattern Mining

Abstract
Social network data provide valuable information for companies to better understand the characteristics of their potential customers with respect to their communities. Yet, sharing social network data in its raw form raises serious privacy concerns because a successful privacy attack not only compromises the sensitive information of the target victim but also divulges the relationship with his/her friends or even their private information. In recent years, several anonymization techniques have been proposed to solve these issues. Most of them focus on how to achieve a given privacy model but fail to preserve the data mining knowledge required for data recipients. In this paper, we propose a method to \(k\)-anonymize a social network dataset with the goal of preserving frequent sharing patterns and maximal frequent sharing patterns, the most important kinds of knowledge required for marketing and consumer behavior analysis. Experimental results on real-life data illustrate the trade-off between privacy and utility loss with respect to the preservation of (maximal) frequent sharing patterns.
Benjamin C. M. Fung, Yan’an Jin, Jiaming Li, Junqiang Liu

A Comprehensive Analysis of Detection of Online Paid Posters

Abstract
We initiate a systematic study to help distinguish a special group of online users, called hidden paid posters, or termed “Internet water army” in China, from the legitimate ones. On the Internet, the paid posters represent a new type of online job opportunities. They get paid for posting comments or articles on different online communities and web sites for hidden purposes, e.g., to influence the opinion of other people toward certain social events or business markets. While being an interesting strategy in business marketing, paid posters may create a significant negative effect on the online communities, since the information from paid posters is usually not trustworthy. When two competitive companies hire paid posters to post fake news or negative comments about each other, normal netizens may feel overwhelmed and find it difficult to put any trust in the information they acquire from the Internet. In this paper, we thoroughly investigate the behavioral pattern of online paid posters based on real-world trace data. We design and validate a new detection mechanism, using both nonsemantic analysis and semantic analysis, to identify potential online paid posters. Our test results with real-world datasets show a very promising performance.
Cheng Chen, Kui Wu, Venkatesh Srinivasan, Xudong Zhang

An Improved Collaborative Recommendation System by Integration of Social Tagging Data

Abstract
Recently a lot of research efforts have been spent on building recommender systems by utilizing the abundant online social network data. In this study, we intend to enhance the recommendation accuracy via integrating social networking information with the traditional recommendation algorithms. To achieve this goal, we first propose a new user similarity metric that not only considers tagging activities of users, but also incorporates their social relationships such as friendship and membership, in measuring the closeness of two users. Then we define a new item prediction method which makes use of both user-to-user similarity and item-to-item similarity. Experimental outcomes on Last.fm data produce the positive results that show the accuracy of our proposed approach.
Sogol Naseri, Arash Bahrehmand, Chen Ding

Personalization of Web Search Using Social Signals

Abstract
Over the last few years, Web has changed significantly. Emergence of social networksSocial network and Web 2.0 have enabled people to interact with Web document in new ways not possible before. In this paper, we present PERSOSE Personalized search engine (PERSOSE) a new search engineSearch engine that personalizes the search results based on users’ social actions.Social actions Although the users’ social actions may sometimes seem irrelevant to the search, we show that they are actually useful for personalization.Personalization We propose a new relevance modelRelevance model called persocial relevance model utilizing three levels of social signals to improve the Web search.Web search We show how each level of persocial model (users’ social actions, friends’ social actions and social expansion) can be built on top of the previous level and how each level improves the search results. Furthermore, we develop several approaches to integrate persocial relevance model into the textual Web search process. We show how PERSOSE Personalized search engine (PERSOSE) can run effectively on 14 million WikipediaWikipedia articles and social data from real FacebookFacebook@Facebook users and generate accurate search results. Using PERSOSE, we performed a set of experiments and showed the superiority of our proposed approaches. We also showed how each level of our model improves the accuracy of search results.
Ali Khodaei, Sina Sohangir, Cyrus Shahabi

The Pareto Principle Is Everywhere: Finding Informative Sentences for Opinion Summarization Through Leader Detection

Abstract
Most previous works on opinion summarization focus on summarizing sentiment polarity distribution toward different aspects of an entity (e.g., battery life and screen of a mobile phone). However, users’ demand may be more beyond this kind of opinion summarization. Besides such coarse-grained summarization on aspects, one may prefer to read detailed but concise text of the opinion data for more information. In this paper, we propose a new framework for opinion summarization. Our goal is to assist users to get helpful opinion suggestions from reviews by only reading a short summary with a few informative sentences, where the quality of summary is evaluated in terms of both aspect coverage and viewpoints preservation. More specifically, we formulate the informative sentence selection problem in opinion summarization as a community leader detection problem, where a community consists of a cluster of sentences toward the same aspect of an entity and leaders can be considered as the most informative sentences of the corresponding aspect. We develop two effective algorithms to identify communities and leaders. Reviews of six products from Amazon.com are used to verify the effectiveness of our method for opinion summarization.
Linhong Zhu, Sheng Gao, Sinno Jialin Pan, Haizhou Li, Dingxiong Deng, Cyrus Shahabi

Social Media Question Asking: A Developing Country Perspective

Abstract
The last decade has seen the emergence of the social networking sites (SNS) and researchers are investigating the useful applications of this technology in various areas apart from its recreational value. Ubiquitous presence of SNS has enabled us to obtain customized information seamlessly from our acquaintance. There have been many works that analyzed the types and topics of questions people ask in these networks and why. Topics like what motivates people to answer such queries, how to integrate the traditional search engines, and SNS together are also well investigated. In this research, we focus on the use of this technology in underdeveloped parts of the world and the new doors it has opened for its inhabitants in terms of obtaining information. Analyzing 880 status messages collected from a widely used SNS, we have observed that, unavailability and inadequacy of information on web in developing countries play a significant role to motivate users using SNS for information retrieval. Based on a structured survey on 328 persons, we have tried to emphasize the differences between social search and traditional web search. Our statistical analysis finds the correlations among different relevant parameters and provides insight that one might require to consider while developing any application for SNS-based searching.
Hasan Shahid Ferdous, Mashrura Tasnim, Saif Ahmed, Md. Tanvir Alam Anik

Evolutionary Influence Maximization in Viral Marketing

Abstract
With the growth of social networks, significant amount of data is brought online that can benefit applications of many kinds if being effectively utilized. As a typical example, Domnigos proposed the concept of viral marketing, which uses the “word of mouth” marketing technique over virtual networks (Domingos, IEEE Intell Syst 20:80–82, 2005). Each user is associated with a network value that represents his/her influence in the network. The network value is used along with other intrinsic features that represent user shopping behaviors for the selection of a small subset of most influential users in the network for marketing purpose. However, most existing viral marketing techniques ignore the dynamic nature of the virtual network where both the features and the relationship of users may change over time. In this paper, we develop a novel framework for the selection of users by exploiting the temporal dynamics of the network. Incorporating temporal dynamics of the network would assist in selecting an optimal subset of users with the maximum influence over the network. This paper focuses on developing an algorithm for the selection of the users to market the product by exploiting the temporal and the structural dynamics of the network. Extensive experimental results over real-world datasets clearly demonstrate the effectiveness of the proposed framework.
Sanket Anil Naik, Qi Yu

Mining and Analyzing the Italian Parliament: Party Structure and Evolution

Abstract
The roll calls of the Italian Parliament in the XVI legislature are studied by employing multidimensional scaling, hierarchical clustering, and network analysis. In order to detect changes in voting behavior, the roll calls have been divided in seven periods of six months each. All the methods employed pointed out an increasing fragmentation of the political parties endorsing the previous government that culminated in its downfall. By using the concept of modularity at different resolution levels, we identify the community structure of Parliament and its evolution in each of the considered time periods. The analysis performed revealed as a valuable tool in detecting trends and drifts of Parliamentarians. It showed its effectiveness at identifying political parties and at providing insights on the temporal evolution of groups and their cohesiveness, without having at disposal any knowledge about political membership of Representatives.
Alessia Amelio, Clara Pizzuti

Backmatter

Weitere Informationen

Premium Partner

    Bildnachweise