LGA: latent genre aware micro-video recommendation on social media

Social media has evolved into one of the most important channels to share micro-videos nowadays. The sheer volume of micro-videos available in social networks often undermines users’ capability to choose the micro-videos that best fit their interests. Recommendation appear as a natural solution to this problem. However, existing video recommendation methods only consider the users’ historical preferences on videos, without exploring any video contents. In this paper, we develop a novel latent genre aware micro-video recommendation model to solve the problem. First, we extract user-item interaction features, and auxiliary features describing both contextual and visual contents of micro-videos. Second, these features are fed into the neural recommendation model that simultaneously learns the latent genres of micro-videos and the optimal recommendation scores. Experiments on real-world dataset demonstrate the effectiveness and the efficiency of our proposed method compared with several state-of-the-art approaches.     

Multimedia summarization using social media content

In this work, we propose a novel multimedia summarization technique from Online Social Networks (OSNs). In particular, we model each Multimedia Social Network (MSN)—i.e. an OSN focusing on the management and sharing of multimedia information—using an hypergraph based approach and exploit influence analysis methodologies to determine the most important multimedia objects with respect to one or more topics of interest. Successively, we obtain from the list of candidate objects a multimedia summary using a summarization model together with an heuristics that aims to generate summaries with priority (with respect to some user keywords), continuity, variety and not receptiveness features. The performed experiments on Flickr shows the effectiveness of proposed approach.     

Quantifying the trustworthiness of social media content

The growing popularity of social media in recent years has resulted in the creation of an enormous amount of user-generated content. A significant portion of this information is useful and has proven to be a great source of knowledge. However, since much of this information has been contributed by strangers with little or no apparent reputation to speak of, there is no easy way to detect whether the content is trustworthy. Search engines are the gateways to knowledge but search relevance cannot guarantee that the content in the search results is trustworthy. A casual observer might not be able to differentiate between trustworthy and untrustworthy content. This work is focused on the problem of quantifying the value of such shared content with respect to its trustworthiness. In particular, the focus is on shared health content as the negative impact of acting on untrustworthy content is high in this domain. Health content from two social media applications, Wikipedia and Daily Strength, is used for this study. Sociological notions of trust are used to motivate the search for a solution. A two-step unsupervised, feature-driven approach is proposed for this purpose: a feature identification step in which relevant information categories are specified and suitable features are identified, and a quantification step for which various unsupervised scoring models are proposed. Results indicate that this approach is effective and can be adapted to disparate social media applications with ease.     

Perceived risks and risk management of social media in an organizational context

The use of new electronic media for marketing communication is gaining in popularity with organizations, and the adoption of social media (SOME) is enjoying particularly rapid growth. However, organizations are uncertain about using SOME and perceived risks are limiting their use. This study explores the key types of risks that business organizations associate with using SOME in their marketing communication. We also explore the effectiveness of procedural control and proactive focusing processes in managing these risks. The results show that organizations perceive three types of risks that deterred companies from increasing the use of SOME. Companies applied procedural control mechanisms to manage time-loss risks. An organization’s familiarity with SOME was found to have a strong effect on time-loss as well as on other types of risks. Research revealed that the role that proactive focus and procedural control played in managing SOME-related risks was less than had been anticipated.     

Exploring virtual classrooms: network communication in a cross-cultural context

Abstract

Using only desktop computers linked to gateway networks, students from around the world can communicate in a ‘virtual classroom'. (A term which is similar in concept to, but in no way associated with the Virtual Classroom Project, a registered trademark of the New Jersey Institute of Technology (Hiltz 1990).) One such network, the Apple Global Education network (AGE), ? ? Developed by Apple Computer Inc., Cupertino, CA, USA. links over one hundred classes from over 20 countries. This network was evaluated with respect to its role in facilitating cross-cultural communication activities.

The nature of communication on the network is described by classifying transmissions sent during a major international crisis: the Gulf War. A taxonomy is proposed containing four levels of network function: audience, diffuser, repository and facilitator. Following the evaluation of the AGE network, recommendations are given to educators regarding network activities in relation to the age and experience of the learner.     

Exploring sharing patterns for video recommendation on YouTube-like social media

YouTube-like video sharing sites (VSSes) have gained increasing popularity in recent years. Meanwhile, Face-book-like online social networks (OSNs) have seen their tremendous success in connecting people of common interests. These two new generation of networked services are now bridged in that many users of OSNs share video contents originating from VSSes with their friends, and it has been shown that a significant portion of views of VSS videos are attributed to this sharing scheme of social networks. To understand how the video sharing behavior, which is largely based on social relationship, impacts users’ viewing pattern, we have conducted a long-term measurement with RenRen and YouKu, the largest online social network and the largest video sharing site in China, respectively. We show that social friends have higher common interest and their sharing behaviors provide guidance to enhance recommended video lists. In this paper, we take a first step toward learning OSN video sharing patterns for video recommendation. An autoencoder model is developed to learn the social similarity of different videos in terms of their sharing in OSNs. We, therefore, propose a similarity-based strategy to enhance video recommendation for YouTube-like social media. Evaluation results demonstrate that this strategy can remarkably improve the precision and recall of recommendations, as compared to other widely adopted strategies without social information.     

Adoption of travel information in user-generated content on social media: the moderating effect of social presence

Travel patterns have gradually changed from group travel to individual travel. An increasing number of people acquire travel information through various types of media. One of the alternative information sources is social media, which enables users to exchange information among members. However, one of the characteristics of social media is information sharing, not information search, which involves both giving (i.e. posting) and taking (i.e. selective reading, forwarding, replying, linking, and liking) information. Compared to the ‘giving’ side of information-sharing research, less effort has been spent on the ‘taking’ side of information research. Therefore, we investigate travel information adoption in social media as well as how individuals communicate with each other. We use the elaboration likelihood model, which measures the impact of central (e.g. argument quality) and peripheral (e.g. credibility) cues on traveller information-sharing behaviour corresponding with social presence on social media. The results of an empirical analysis of 527 respondents, who were experienced in travel information adoption via social media, were examined. Our findings revealed that argument quality had a positive effect on perceived usefulness and source credibility positively affected perceived usefulness and social relationships. Perceived usefulness had a significant positive effect on social relationships. Both perceived usefulness and social relationships affected travel information adoption. Lastly, the levels of argument quality and source credibility perceived by social media members were found to differ according to the level of social presence.     

UR: SMART–A tool for analyzing social media content

Information Systems and e-Business Management - The digital transformation, with its ongoing trend towards electronic business, confronts companies with increasingly growing amounts of data which...     

Exploring the conformational space of cyclic peptides by a stochastic search method

A stochastic search algorithm is applied in order to probe the conformations of cyclic peptides. The search is conducted in two stages. In the first stage, random conformations are generated and evaluated by a penalty function for ring closure ability, following a stepwise construction of each amino acid into the peptide by a random choice of one of its allowed conformations. The allowed conformational ranges of backbone dihedral angles for each amino acid have been extracted from a Data Bank of diverse proteins. Values of dihedral angles that do not contribute favorably to the scoring of ring closure are retained or discarded by a statistical test. Values are discarded up to a point from which all remaining combinations of angles are constructed, scored, sorted, and clustered. In the second stage, side chains have been added and fast optimization was applied to the set of diverse conformations in a "united atoms" approach, with the "Kollman forcefield" of Sybyl 6.8. This iterative stochastic elimination algorithm finds the global minimum and most of the best results, when compared to a full exhaustive search in appropriately sized problems. In larger problems, we compare the results to experimental structures. The root mean square deviation (RMSD) of our best results compared to crystal structures of cyclic peptides with sizes from 4 to 15 amino acids are mostly below 1.0 A up to 8 mers and under 2.0 A for larger cyclic peptides.     

FCE-SVM: a new cluster based ensemble method for opinion mining from social media

Opinion mining aiming to automatically detect subjective information has raised more and more interests from both academic and industry fields in recent years. In order to enhance the performance of opinion mining, some ensemble methods have been investigated and proven to be effective theoretically and empirically. However, cluster based ensemble method is paid less attention to in the area of opinion mining. In this paper, a new cluster based ensemble method, FCE-SVM, is proposed for opinion mining from social media. Based on the philosophy of divide and conquer, FCE-SVM uses fuzzy clustering module to generate different training sub datasets in the first stage. Then, base learners are trained based on different training datasets in the second stage. Finally, fusion module is employed to combine the results of based learners. Moreover, the multi-domain opinion datasets were investigated to verify the effectiveness of proposed method. Empirical results reveal that FCE-SVM gets the best performance through reducing bias and variance simultaneously. These results illustrate that FCE-SVM can be used as a viable method for opinion mining.     

Enriching one-class collaborative filtering with content information from social media

In recent years, recommender systems have become popular to handle the information overload problem of social media websites. The most widely used Collaborative Filtering methods make recommendations by mining users’ rating history. However, users’ behaviors in social media are usually implicit, where no ratings are available. This is a One-Class Collaborative Filtering (OCCF) problem with only positive examples. How to distinguish the negative examples from missing data is important for OCCF. Existing OCCF methods tackle this by the statistical properties of users’ historical behavior; however, they ignored the rich content information in social media websites, which provide additional evidence for profiling users and items. In this paper, we propose to improve OCCF accuracy by exploiting the social media content information to find the potential negative examples from the missing user-item pairs. Specifically, we get a content topic feature for each user and item by probabilistic topic modeling and embed them into the Matrix Factorization model. Extensive experiments show that our algorithm can achieve better performance than the state-of-art methods.     

Customer engagement and performance in social media: a managerial perspective

Electronic Markets - The topic of customer engagement via social media is receiving increasing consideration in the literature. Previously, scholars’ use of the notion and dimensionality of...     

The case analysis on sentiment based ranking of nodes in social media space

Now-a-days, social network sites have become quite popular for communication in the society. People have entangled their day-to-day activities around social media platforms. Social Networks have allowed the users to share their opinions on different topics. In social media, sentiment analysis is an important character to determine opinions of users. Moreover, user’s can be ranked to determine their relative influence. This paper proposes a methodology to rank the users involving sentiment related parameters such as likes, comments and corresponding likescount. Analysis of users’ comments is carried-out. Weights are assigned to these parameters and scores are calculated for each user. Users are ranked on the basis of scores obtained and compared with existing technique. In order to verify the effectiveness of proposed methodology, data is extracted from a verified Facebook page ‘Panjab University, Chandigarh’. Mean, standard deviation and variance are computed to capture the usefulness of ranks obtained by the proposed method. Results depict that the proposed methodology is better than existing technique since it incorporates several features indicating positive and negative behavior of users. This technique can be used to determine the highly trusted and the most distrusted users in a social media user’s profile. Users with negative scores can be considered for outlier analysis. The proposed methodology can also be extended to work on other social media platforms.     

Generational differences in content generation in social media: The roles of the gratifications sought and of narcissism

This study examines the roles of the gratifications sought and of narcissism in content generation in social media and explores the generational differences in motivations and in narcissistic personalities when predicting the usage of Facebook, blogs, and forums. Data were gathered from a probability sample of 596 social media users through a telephone survey in 2010. Factor analysis results showed that content generation using social media was satisfying five socio-psychological needs: showing affection, venting negative feelings, gaining recognition, getting entertainment, and fulfilling cognitive needs. In particular, people who used social media to meet their social needs and their need for affection tended to use Facebook and blogs. In contrast, when users wanted to air out discontent, they often turned to forums. Results also showed that exhibitionists seemed to use social media to show affection, express their negative feelings, and achieve recognition. The study found no generational differences in using Facebook and blogs as a means to satisfy social needs or the need for affection. However, differences in patterns of social media usage were found among Baby Boomers with different narcissistic personalities. The paper includes a discussion of the study’s limitations and suggestions for future research.     

Representing and applying knowledge for argumentation in a social context

The concept of argumentation in AI is based almost exclusively on the use of formal, abstract representations. Despite their appealing computational properties, these abstractions become increasingly divorced from their real world counterparts, and, crucially, lose the ability to express the rich gamut of natural argument forms required for creating effective text. In this paper, the demands that socially situated argumentation places on knowledge representation are explored, and the various problems with existing formalisations are discussed. Insights from argumentation theory and social psychology are then adduced as key contributions to a notion of social context which is both computationally tractable and suitably expressive for handling the complexities of argumentation found in natural language.