Organization, Government and Individual (OGI) have popularized the use of Digital Social Networks (DSN) that reduces the processing time of social-aware tasks. To accomplish a community-based communication, each social-aware task should identify its community group. The identified group uses a task to avail all the DSN benefits to their customers / citizens. As a result, the community-based detection algorithm has played a significant role in literature. However, the existing algorithms have had several challenging issues, such as performance and scalability. Thus, a distributed community detection algorithm is presented using Apache Spark’s Resilient Distributed Data Set (RDD) framework based on the Scala programming language. The Apache Spark framework provides an ideal solution that offers ease of coding, performance, interactive mode and disk Input-Output bottlenecks in Hadoop /Map Reduce. Besides, it presents a platform of distributed community detection that reduces the computational computation by applying transformations, aggregations and joins. The experimental results show that the proposed framework achieves high accuracy for both real-world and synthetic networks.

Sentiment analysis of scientific citations is a novel and remarkable research area. Most of the work on opinion or sentiment analysis has been suggested on social platforms such as Blogs, Twitter, and Facebook. Nevertheless, when it comes to recognizing sentiments from scientific citation papers, investigators used to face difficulties due to the implied and unseen natures of sentiments or opinions. As the citation references are reflected implicitly positive in opinion, famous ranking and indexing prototypes frequently disregard the sentiment existence while citing. Hence, in the proposed framework the paper emphasizes the issue of classifying positive and negative polarity of reference sentiments in scientific research papers. First, the paper scraps the PDF articles from arxiv.org under the computer science group consisting of articles that are comprised of ‘autism’ in their title, then the paper extracted cited references and assigns polarity scores to each cited reference. The paper uses a supervised classifier with a combination of significant feature sets and compared the performance of the models. Experimental results show that a combined CNN-LSTM deep neural network model results in 85% of accuracy while traditional models result in less accuracy.

Of late, text and sentiment analysis have become essential parts of modern marketing. These play a vital role in the division of natural language processing (NLP). It mainly focuses on text classification to examine the intention of the processed text; it can be of positive or negative types. Sentiment analysis dealt with the computational treatment of sentiments, opinions, and subjectivity of text. This chapter tackles a comprehensive approach for the past research solutions that includes various algorithms, enhancements, and applications. This chapter primarily focuses on three aspects. Firstly, the authors present a systematic review of recent works done in the area of text and sentiment analysis; second, they emphasize major concepts, components, functionalities, and classification techniques of text and sentiment analysis. Finally, they provide a comparative study of text and sentiment analysis on the basis of trending research approaches. They conclude the chapter with future directions.

Recommender systems inherently dynamic in nature and exponentially grow with time, in terms of interests and behaviour patterns. Traditional recommender systems rely on similarity of users or items in static networks where the user/item neighbourhood is almost same and they generate the same recommendations since the network is constant. This paper proposes a novel architecture, called Temporal Community-based Collaborative filtering, which is an association of recommendation and the dynamic community algorithm in order to exploit the temporal changes in the community structure to enhance the existing system. Our framework also provides solutions to common inherent issues of collaborative filtering approach such as cold-start, sparsity and compared against static and traditional collaborative systems. The outcomes indicate that the proposed system yields higher values in quality standards and minimizes the drawbacks of the traditional recommender system.

No abstract