Social Bookmarking Services Research Articles

Metaheuristic Ant Lion and Moth Flame Optimization-Based Novel Approach for Automatic Detection of Hate Speech in Online Social Networks

In the online social networks, blogs, microblogs, social bookmarking services and sharing sites, and various web forum pages; the sharing of knowledge, opinions, ideas, etc. are spreading very quickly. This situation brings very dangerous problems in social networks. One of these problems is hate speech detection (HSD) problem which is covering issues such as insults, swearing, humiliation, discrimination, exclusion, detest, abhor, blast, damn, and intolerance. These can be reactions to a person, a group, an organization, an order, or an event. Although few machine learning methods have been used in the literature to solve this important problem in online social media, the performance of the HSD models in terms of many metrics needs to be increased. In this study, an automatic HSD system based on metaheuristic methodology was proposed for better results in this new and important problem. In the proposed optimization approach, Ant Lion Optimization (ALO) algorithm and Moth Flame Optimization (MFO) algorithm were designed for the HSD problem. This is the first attempt to use optimization algorithms as solution search strategies for automatic HSD. An efficient representation scheme and flexible fitness function were designed for this purpose. Many metrics can easily be embedded into the designed fitness function in order to be simultaneously optimized. Firstly, the basic natural language processing (NLP) steps were carried out. Feature extraction was performed using Bag of Words (BoW), Term Frequency (TF), and document vector (Word2Vec). Then, the performances of the proposed novel approaches were analyzed in detail on the three different real-world data. The obtained results were also checked against eight popular supervised machine learning algorithms, Social Spider Optimization (SSO) algorithm, and state-of-the-art Tunicate Swarm Algorithm (TSA). Considering the evaluation criteria for three sets of experiments, it was observed that the accuracy, sensitivity, precision, and f-score results of the ALO and MFO algorithms were superior to machine learning methods. As a result of the experimental studies, the highest accuracy value was 92.1% for ALO, while this value was 90.7% for MFO. Other numerical values obtained in the study were given in the experiments and results section with tables and graphics in detail. Due to the promising results of the proposed approaches, they are anticipated to be used in the solution of many social media and networking problems.

Gender Differences in the Awareness and Use of Social Bookmarking Sites by Faculty of Engineering Colleges in Puducherry, India

In the advent of online technologies, most of the information resources available in the form of digital. So the Faculty may be able to replace or supplement printed information sources with social bookmarking sites. This study reported awareness and use of social bookmarking services by faculty of engineering colleges at Puducherry union territories. The survey research design was employed and questionnaire was used to collect data and SPSS statistical tools used for analyzing the data. This study found that the faculty are largely (70.5%) aware of social bookmarking services. Their familiarity with and usage of the services were very low as very little proportions of the respondents from female faculty members. Most of the proposed null hypotheses are accepted.

Methodical aspects of using of Web 2.0 services in the process of blended learning

The article is devoted to the issues of organization of blended learning using Web 2.0 services intended at providing educational services via the Internet. The learning process is designed by implementing learning technologies based on the active work of students, including cooperation, joint creativity and the development of personally significant Internet resources. At the same time, students interact with the instructor in an active form both internally and remotely. The effectiveness of mastering the course by students in the process of blended learning depends on the ratio of the forms of study used (classroom work, independent work (individual, group), distance interaction) and the nature of the information presented for classroom and independent work. The problem of finding the optimal combination of forms, methods and means of training is one of the key to providing a holistic learning process. The purpose of this work is to consider the methodological aspects of the organization of blended learning using Web 2.0 services to improve the quality of educational services and motivate students to perform various types of educational activities. The main part of the article considers with the educational opportunities of the most popular Web 2.0 services, the pedagogical features and the nature of the implementation of study assignments in the process of blended learning, identifies the specifics of implementing joint activities in the Internet, and determines the approaches to organizing the network interaction of the teacher and students. Web 2.0 services were used to organize open access to the information resources of the course and the personal information space of students, to discuss issues collectively and to collaborate with students on the training project, to provide prompt and reliable communication between students and the teacher, passing tests and interviews to identify the intermediate level knowledge or to determine the attitude of students to some problem. As a theoretical basis for the organization and maintenance of blended learning using Web 2.0 services was used a five-step model of moderating online learning by J. Salmon (access and motivation, online socialization, information exchange, knowledge building, development). The article gives an example of the practical implementation of J. Salmon’s model using Web 2.0 services (blogs, social photo and video services, social bookmarking services, social networking services, etc.). After completion of the training, students were questioned (32 participants) in order to establish their attitude to the use of Web 2.0 services in the learning process. As a result of the performed research, the effectiveness of the proposed methodological approaches to the organization of blended learning has been confirmed. The analysis of the obtained results showed that the process of network interaction is determined by the high level of students’ interest in this activity, the social involvement of students in it, the desire to present their point of view on the problem that has arisen, to justify and prove it to other members of the discourse.

Exploiting Social Annotations to Generate Resource Descriptions in a Distributed Environment: Cooperative Multi-Agent Simulation on Query-Based Sampling

In distributed information retrieval, resource descriptions play a principal role in facilitating the task of other processes such as the resource selection process and the merging process. The previous approach for acquiring resource descriptions was based on different techniques to improve the retrieval process, but they have many limitations. In this paper, we describe a new approach for acquiring precise resource descriptions, based on social annotations available in the social bookmarking service.

Implementation of Folksonomy Based Tag Cloud Model for Information Retrieval from Document Repository in an Indian University

In the magnitude of internet one need to devote extra time to investigate anticipated resource, especially when one need to search information from documents. For the higher range internet there is serious need to demand the essentiality to discover the reserved resources. One of the solutions for information retrieval from document repository is to attach tags to documents. Numerous online social bookmarking services permit users to attach tags with resources which are eventually meta-data, frequently stated as folksonomy. In current paper, authors implemented this model for information retrieval by utilizing these tags, after retrieving by using delicious API and synthesize tag cloud in an Indian University to search and retrieve information from document repository.

Does content categorization lead to knowledge building? An experiment in a social bookmarking service

Designers of systems with knowledge building purposes (learning platforms, serious games, etc.) are increasingly, sometimes inadvertently, using strategies like collaboration (Wikis), social networking, and Gamification. They aim to engage people and help them evolve their knowledge through several forms of interaction (sharing, commenting, editing, liking, replying, tagging, etc.). Given this scenario, the present work analyzed the impact of a collaborative, tag-based content categorization mechanism in user’s perception of knowledge building in a social bookmarking service. To achieve this we performed a statistical analysis based on usage log data recorded from the tagging interactions and a post-usage online survey with 84 responses. Although the user interaction with content (tagging) would suggest higher levels of learning, the results showed that the perception of knowledge building is not affected by user participation behavior, and lurkers and posters could equally benefit from the service.

Exploiting social bookmarking services to build clustered user interest profile for personalized search

Search engine users tend to write short queries, generally comprising of two or three query words. As these queries are often ambiguous or incomplete, search engines tend to return results whose rankings reflect a community of intent. Moreover, search engines are designed to satisfy the needs of the general populace, not those of a specific searcher. To address these issues, we propose two methods that use Singular Value Decomposition (SVD) to build a Clustered User Interest Profile (CUIP), for each user, from the tags annotated by a community of users to web resources of interest. A CUIP consists of clusters of semantically or syntactically related tags, each cluster identifying a topic of the user’s interest. The matching cluster, to the given user’s query, aids in disambiguation of user search needs and assists the search engine to generate a set of personalized search results. A series of experiments was executed against two data sets to judge the clustering tendency of the cluster structure CUIP, and to evaluate the quality of personalized search. The experiment results indicate that the CUIP based personalized search outperforms the baseline search and is better than the other approaches that use social bookmarking services for building a user profile and use it for personalized search.

An Efficient Method for Noisy Annotation Data Modeling

Probabilistic topic models are used for analyzing and extracting content-related annotations from noisy annotated discrete data like WebPages on WWW and these WebPages are stored using social bookmarking services with the help of social bookmarking services, reason behind this process most of time users can attach annotations freely, some annotations do not describe the semantics of the content, therefore they are noisy, simply they are not content related. The extraction of content-related annotations can be used as a prepossessing step in machine learning. Prepossessing step in machine learning is like text classification and image recognition, and can improve information retrieval performance. The proposed model is a generative model for content and annotations, where annotations are assumed to be originated either from topics that generated the content or from a general distribution unrelated to the content. We demonstrate the effectiveness of the proposed method with the help of synthetic data and real social annotation data for text and images.

個人ごとのタグ空間の類似性に基づく協調フィルタリング手法

This paper improves Collaborative Filtering Algorithms (CF) to recommend similar bookmarks for users of Social Bookmark Services. Standard CF algorithms have a problem with precision in multi-genre cases such as social bookmarks. We aim at solving this issue and propose a personal classification space-based collaborative filtering algorithm. Our interpretation is that users' bookmarks are disposed in their own classification space made by tags. The bookmark is transformed into a scalar as a similarity to a starting bookmark for a recommendation. We compare the personal classification space with others' space using the similarities. We handle the similarities as the rating to the bookmark in standard CF algorithms. The proposed algorithm is compared with other existing algorithms. In conclusion, our algorithm shows better precision than others.

Academic social bookmarking: An empirical analysis of Connotea users

Three groups of social bookmarking users, namely experienced users, users with moderate level of experience, and novice users, were investigated, in terms of their reported tagging behaviors, perceived usefulness of social bookmarking in information discovery and management, and perceived usefulness of the bookmarking features provided. Based on the empirical analysis of the Connotea users, who are primarily in academia, the study shows that experienced academic users generally prefer to use social bookmarking while moderately experienced and novice users still prefer to use the traditional bookmarking methods, such as creating and using bookmarks on a dedicated computer. Experienced academic users were also found to create more tags per bookmark comparing to the other two groups. Most novice academic users, however, only created one tag per bookmark, which just met Connotea's minimum requirement. Surprisingly different from the collaborative design nature of the social bookmarking systems, the study finds that our participants, particularly experienced academic users, prefer to create and use their own bookmarks rather than sharing bookmarks created by others. In fact, experienced users have significantly higher frequency (once every two weeks) of creating bookmarks than the other two groups (once a month or less). In addition, website design features and functions, such as automatic collection of bibliographic information, are regarded by all participants as helpful for information discovery. The in-depth examination and discussion of the opinions of Connotea users may be useful for further improvement of the design features and usage applications, particularly for academic social bookmarking websites. The results may also have potential implications to the future development of social bookmarking services in general.

Social semantic query expansion

Weak semantic techniques rely on the integration of Semantic Web techniques with social annotations and aim to embrace the strengths of both. In this article, we propose a novel weak semantic technique for query expansion. Traditional query expansion techniques are based on the computation of two-dimensional co-occurrence matrices. Our approach proposes the use of three-dimensional matrices, where the added dimension is represented by semantic classes (i.e., categories comprising all the terms that share a semantic property) related to the folksonomy extracted from social bookmarking services, such as delicious and StumbleUpon . The results of an indepth experimental evaluation performed on both artificial datasets and real users show that our approach outperforms traditional techniques, such as relevance feedback and personalized PageRank, so confirming the validity and usefulness of the categorization of the user needs and preferences in semantic classes. We also present the results of a questionnaire aimed to know the users opinion regarding the system. As one drawback of several query expansion techniques is their high computational costs, we also provide a complexity analysis of our system, in order to show its capability of operating in real time.

Modeling Noisy Annotated Data with Application to Social Annotation

We propose a probabilistic topic model for analyzing and extracting content-related annotations from noisy annotated discrete data such as webpages stored using social bookmarking services. With these services, because users can attach annotations freely, some annotations do not describe the semantics of the content, thus they are noisy, i.e., not content related. The extraction of content-related annotations can be used as a prepossessing step in machine learning tasks such as text classification and image recognition, or can improve information retrieval performance. The proposed model is a generative model for content and annotations, in which the annotations are assumed to originate either from topics that generated the content or from a general distribution unrelated to the content. We demonstrate the effectiveness of the proposed method by using synthetic data and real social annotation data for text and images.

Web clustering using social bookmark data regarding user network

AbstractSocial bookmark services are Web services which enable users to share their personal bookmark data with other users. In our research, we propose a web clustering method in a social bookmark service using bookmark data modified with user network data. We define bookmark data as pairs consisting of a user and a web page with bookmarks, and user network data as links between users and their favorite users. In more detail, we use probabilistic latent semantic indexing (PLSI), which makes it possible to project social bookmark data on a latent semantic space and classify them by their semantic attributes. Using PLSI, we can classify web pages according to their topics, and for a user network, we can append all bookmark data of users who are reachable by a given user to that user's bookmark data. To confirm the effectiveness of our proposed clustering method, we evaluated a clustering examination using real social bookmark data. The results confirmed that our proposed method is capable of producing comparable results in the classification of web pages by topic. © 2013 Wiley Periodicals, Inc. Electron Comm Jpn, 96(3): 24–30, 2013; Published online in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/ecj.11450

Personal classification space-based collaborative filtering algorithms

This research proposes a new similar information recommendation system focusing on social bookmarking, which organises bookmarks by using tags. Since social bookmarking targets a wide variety of genres of web pages, this research handles the problem where the standard collaborative filtering method cannot offer recommendations with a better level of precision. This paper improves the collaborative filtering algorithm for users of social bookmark services. A user's bookmarks are placed on his/her own classifying space made of tags. These bookmarks are transformed into a degree of similarity for recommendations. The degree is used to compare the personal classifying space with another's space. Comparison with previous studies confirms the superiority of the method based on space classification, in particular, where the cos distance with the distribution weight added is used as similarity between items. This proposed method shows a significant superiority.

Using a Network Analysis Approach for Organizing Social Bookmarking Tags and Enabling Web Content Discovery

This article describes an innovative approach to reorganizing the tag space generated by social bookmarking services. The objective of this work is to enable effective search and discovery of Web content using social bookmarking tags. Tags are metadata generated by users for Web content annotation. Their potential as effective Web search and discovery tool is hindered by challenges such as, the tag space being untidy due to ambiguity, and hidden or implicit semantics. Using a novel analytics approach, we conducted network analyses on tags and discovered that tags are generated for different purposes and that there are inherent relationships among tags. Our approach can be used to extract the purposes of tags and relationships among the tags and this information can be used as facets to add structure and hierarchy to reorganize the flat tag space. The semantics of relationships and hierarchy in our proposed faceted model of tags enable searches on annotated Web content in an effective manner. We describe the implementation of a prototype system called FASTS to demonstrate feasibility and effectiveness of our approach.

Expert vs. Novices Dimensions of Tagging Behaviour in an Educational Setting

The organization and representation of information and knowledge have always been exclusively in the domain of professionals and experts. This has begun to change with the development of folksonomies as alternative, user-generated models of organizing information. The aim of this paper is to research the efficiency in tagging and folksonomy. The flexibility of tagging allows users to classify their collections of items in the ways that they find useful, but the personalized variety of terms can present challenges when searching and browsing. In order to determine the efficiency of tagging research evidence about the nature of tagging and tagging behaviour of specific user groups is needed. This paper contributes to research findings in this domain by presenting findings from a study exploring differences in expert and novices tagging. The research was conducted by giving freshman students, with no prior knowledge of tagging or indexing and therefore determined as novices, an article in the social bookmarking service Delicious. Based only on title, subtitle and abstract of the article every student was supposed to assign tags to that article and do the same after reading the whole article. The same procedure was repeated with postgraduate students from the Department of Information Sciences with sufficiently experience and knowledge in tagging and indexing. In this way differences or similarities between tagging by more advanced users and tagging by average / amateur users could be analyzed and compared. The research has surfaced differences in tag numbers and tag distributions. The findings indicate more precision and consistency in tagging of the expert group, indicating that education in tagging could raise the quality of folksonomies on the long term.

Measuring researchers’ use of scholarly information through social bookmarking data: A case study of BibSonomy

This paper explores the possibility of using data from social bookmarking services to measure the use of information by academic researchers. Social bookmarking data can be used to augment participative methods (e.g. interviews and surveys) and other, non-participative methods (e.g. citation analysis and transaction logs) to measure the use of scholarly information. We use BibSonomy, a free resource-sharing system, as a case study. Results show that published journal articles are by far the most popular type of source bookmarked, followed by conference proceedings and books. Commercial journal publisher platforms are the most popular type of information resource bookmarked, followed by websites, records in databases and digital repositories. Usage of open access information resources is low in comparison with toll access journals. In the case of open access repositories, there is a marked preference for the use of subject-based repositories over institutional repositories. The results are consistent with those observed in related studies based on surveys and citation analysis, confirming the possible use of bookmarking data in studies of information behaviour in academic settings. The main advantages of using social bookmarking data are that is an unobtrusive approach, it captures the reading habits of researchers who are not necessarily authors, and data are readily available. The main limitation is that a significant amount of human resources is required in cleaning and standardizing the data.

Web Clustering Using Social Bookmark Data Regarding User Network

Social Bookmark services are web services which enable users to share their personal bookmark data with other users. In our research, we propose a web clustering method in a social bookmark service using bookmark data modified with user network data. Here, we define bookmark data as a pair of a user and a webpage he bookmarks, and user network data as links between users and their favorite users. In more detail, we use Probabilistic Latent Semantic Indexing (PLSI) which enables to project social bookmark data on latent semantic space and classify them by their semantic attributes. Using PLSI, we can classify webpages according to their topics, and regarding user network, we can append all bookmark data of user, who are reachable for a user, to his bookmark data. To confirm an effectiveness of our proposed clustering method, we have evaluated a clustering examination using real Social Bookmark Data. And finally we have confirmed that our proposed method could have classified webpages in the same topic.

From Learning Style of Webpage Content to Learner's Learning Style

Technology plays an important role in the development of students who can search for the concepts which they learn in the books on the Internet and find out more information on them. This will increase the depth of their knowledge. According to some researches, students tend to be more active and more participative when technology is being integrated in their lesson resulting to better comprehension and good performance. Using technology in the learning process can facilities automatic detection of the learner’s learning styles which can help the learner to develop his coping strategies to compensate for his/her weaknesses, capitalize on his/her strengths, improve the quality of the learning process and make it more effective. This research presents an automatic tool for detecting learning styles in a learning environment by analyzing the content of the learner’s favorite WebPages using social bookmarking services(www.tagme1.com) and shows that how actual behavior of the learners during the learning process can be used as an effective source for detecting their learning styles based on Felder-Silverman learning style model (FSLSM) .

Distributed Social Bookmarking Web Service Architecture. SOAP vs. iCamp FeedBack

Social bookmarking services became very popular recently. Easy of use, possibility to share and discover in addition to accessibility though the Internet, turns social bookmarking sys- tems into powerful repository of shared knowledge. Obviously this attracts attention of educational institutions and recently such systems started to appear under their domains. However, usually these systems stay separate and limit their users by their bounds. It means that separate systems' students could reach each other and use knowledge base, aggregated in other systems. On the other side, institutions usually want to own this assembled data and do not give away collected knowledge base to third side. This issue does not allow building social bookmarking systems that can be used by multi-institutional users. An idea is to develop distributed system where every institution will have their own database, but, on the other hand, will allow exploring and using data from other net- work sources. This article overviews possible distributed system architecture models and suggests a solution that will eliminate such service issues. Moreover, two different approaches towards dis- tributed services communication are evaluated in this article: SOAP vs. iCamp FeedBack. SOAP is a lightweight XML based protocol for exchanging structured information between distributed applications. FeedBack is another model that uses plain RSS feed to transmit data. Both models are tested and evaluated in this article.