Detecting social bots and identifying social botnet communities are extremely important in online social networks (OSNs). In this paper, we first construct a weighted signed Twitter network graph based on the behavioral similarity and trust values between the participants (i.e., OSN accounts) as weighted edges. The behavioral similarity is analyzed from the viewpoints of tweet-content similarity, shared URL similarity, interest similarity, and social interaction similarity for identifying similar types of behavior (malicious or not) among the participants in the Twitter network; whereas the participant's trust value is determined by a random walk model. Next, we design two algorithms - Social Botnet Community Detection (SBCD) and Deep Autoencoder based SBCD (called DA-SBCD) - where the former detects social botnet communities of social bots with malicious behavioral similarity, while the latter reconstructs and detects social botnet communities more accurately in presence of different types of malicious activities. Finally, we evaluate the performance of proposed algorithms with the help of two Twitter datasets. Experimental results demonstrate the efficacy of our algorithms with better performance than existing schemes in terms of normalized mutual information (NMI), precision, recall and F-measure. More precisely, the DA-SBCD algorithm achieves about 90% precision and exhibits up to 8% improvement on NMI.

Social media data has become invaluable component of business analytics. A multitude of nuances of social media text make the job of conventional text analytical tools difficult. Code-mixing of text is a phenomenon prevalent among social media users, wherein words used are borrowed from multiple languages, though written in the commonly understood roman script. All the existing supervised learning methods for tasks such as Parts Of Speech (POS) tagging for code-mixed social media (CMSM) text typically depend on a large amount of training data. Preparation of such large training data is resource-intensive, requiring expertise in multiple languages. Though the preparation of small dataset is possible, the out of vocabulary (OOV) words pose major difficulty, while learning models from CMSM text as the number of different ways of writing non-native words in roman script is huge. POS tagging for code-mixed text is non-trivial, as tagging should deal with syntactic rules of multiple languages. The important research question addressed by this article is whether abundantly available unlabeled data can help in resolving the difficulties posed by code-mixed text for POS tagging. We develop an approach for scraping and building word embeddings for code-mixed text illustrating it for Bengali-English, Hindi-English, and Telugu-English code-mixing scenarios. We used a hierarchical deep recurrent neural network with linear-chain CRF layer on top of it to improve the performance of POS tagging in CMSM text by capturing contextual word features and character-sequence–based information. We prepared a labeled resource for POS tagging of CMSM text by correcting 19% of labels from an existing resource. A detailed analysis of the performance of our approach with varying levels of code-mixing is provided. The results indicate that the F1-score of our approach with custom embeddings is better than the CRF-based baseline by 5.81%, 5.69%, and 6.3% in Bengali, Hindi , and Telugu languages, respectively.

Malicious social bots generate fake tweets and automate their social relationships either by pretending like a follower or by creating multiple fake accounts with malicious activities. Moreover, malicious social bots post shortened malicious URLs in the tweet in order to redirect the requests of online social networking participants to some malicious servers. Hence, distinguishing malicious social bots from legitimate users is one of the most important tasks in the Twitter network. To detect malicious social bots, extracting URL-based features (such as URL redirection, frequency of shared URLs, and spam content in URL) consumes less amount of time in comparison with social graph-based features (which rely on the social interactions of users). Furthermore, malicious social bots cannot easily manipulate URL redirection chains. In this article, a learning automata-based malicious social bot detection (LA-MSBD) algorithm is proposed by integrating a trust computation model with URL-based features for identifying trustworthy participants (users) in the Twitter network. The proposed trust computation model contains two parameters, namely, direct trust and indirect trust. Moreover, the direct trust is derived from Bayes' theorem, and the indirect trust is derived from the Dempster-Shafer theory (DST) to determine the trustworthiness of each participant accurately. Experimentation has been performed on two Twitter data sets, and the results illustrate that the proposed algorithm achieves improvement in precision, recall, F-measure, and accuracy compared with existing approaches for MSBD.

Most of the oil industries have a substantial volume of physical subsurface data generated as part of the exploration study. This data is collected over many decades and exists in various formats such as tapes, cartridges, CDs, DVDs, paper media comprising of maps, technical well reports, and seismic logs. These items are usually stored in large offsite repositories across the globe and is maintained by third-party vendors. Access to this historical data is crucial for oil companies as it helps to find potential prospects for oil extraction which otherwise require an exploratory study by geologists using satellite imagery, surface rocks, terrain, and seismology. Storing large volumes of technical data in offsite repositories also posts many key challenges such as high storage cost, high retrieval time and inaccessibility of information. To address the above challenges, companies are digitizing the physical data and complementing with rich metadata extraction by Optical Character Recognition(OCR). This introduces some more technical challenges while dealing with lower Dots Per Inch (DPI) scans, poor quality scans, and huge file size. Several frameworks are developed which store the data in local repositories but these frameworks have limitations with respect to the number of documents processed, huge file size and storage scalability. To deal with above-mentioned problems, we present a high-performance computing cloud-based framework by storing the digitized data in the cloud, metadata enrichment through OCR along with image enhancement by a series of Image Processing (IP) techniques and provide high data availability to users using cloud-based search. We have tested this framework with big oil and gas company's data on a huge scale and the results are encouraging. Although this paper addresses oil industries domain problem, the proposed framework can be applied to other domains that have huge physical data.

Community structures in networks are helpful to understand the network structure and analyze the network properties. Majority of the studies in this area tend to discover communities by analyzing the linkages in the networks, which may result in communities with diversified attributes. In this paper, we model community detection as an optimization problem. This paper proposes a novel community detection approach by leveraging the concepts of particle swarm optimization with dynamic neighborhood topology by analyzing the similarity between the node attributes. The process starts by dividing the particles into several sub swarms, and the particles in each sub swarm iteratively form communities through information exchange and integrating the nodes which satisfy a threshold similarity score. The presented method does not require any prior knowledge about the communities. Experiments are carried out on real-life datasets, and the process is executed in parallel, asynchronous environments. We evaluated our method with the evaluation properties namely Omega index, Silhouette coefficient and Tanimoto coefficient to evaluate the performance and quality of the clusters. We also compared our approach with existing approaches, and the results demonstrate that our method performed better in terms of measured metrics.

Recommendation systems play an important role in suggesting relevant information to users. In this paper, we introduce community-wise social interactions as a new dimension for recommendations and present a social recommendation system using collaborative filtering and community detection approaches. We use (i) community detection algorithm to extract friendship relations among users by analyzing user-user social graph and (ii) user-item based collaborative filtering for rating prediction. We developed our approach using map-reduce framework. Our approach improves scalability, coverage and cold start issue of collaborative filtering based recommendation system. We carried out experiments on MovieLens and Facebook datasets, to predict the rating of the movie and produce top-k recommendations for new (cold start) user. The results are compared with traditional collaborative filtering based recommendation system.

Detecting communities is of great importance in various disciplines such as social media, biology and telephone networks, where systems are often represented as graphs. Community is formed by individuals such that those within a group interact with each other more frequently than with those outside the group. The communities have different properties such as node degree, betweenness, centrality, cluster coefficient and modularity. Discovering communities from social networks of big data scale on a single se quential machine is a tedious task. In this paper, we present a Scalable Community Detection Algorithm which relaxes the performance issues due to many I/Os. We adopt Girvan-Newman's modularity based hierarchical community detection algorithm in bottom u p a pproach an d proposed an approximation algorithm for community detection in a distributed environment. We developed our approach using MapReduce and Giraph computing platforms. Experimental results demonstrate that the proposed approach is more efficient than standard MapReduce approach and easily scaled to graph of any size.

Cloud computing has taken the limelight with respect to the present industry scenario due to its multi-tenant and pay-as-you-use models, where users need not bother about buying resources like hardware, software, infrastructure, etc. on an permanently basis. As much as the technological benefits, cloud computing also has its downside. By looking at its financial benefits, customers who cannot afford initial investments, choose cloud by compromising on its concerns, like security, performance, estimation, availability, etc. At the same time due to its risks, customers - relatively majority in number, avoid migration towards cloud. Considering this fact, performance and estimation are being the major critical factors for any application deployment in cloud environment; this paper brings the roadmap for an improved performance-centric cloud storage estimation approach, which is based on balanced PCTFree allocation technique for database systems deployment in cloud environment. Objective of this approach is to highlight the set of key activities that have to be jointly done by the database technical team and business users of the software system in order to perform an accurate analysis to arrive at estimation for sizing of the database. For the evaluation of this approach, an experiment has been performed through varied-size PCTFree allocations on an experimental setup with 100000 data records. The result of this experiment shows the impact of PCTFree configuration on database performance. Basis this fact, we propose an improved performance-centric cloud storage estimation approach in cloud. Further, this paper applies our improved performance-centric storage estimation approach on decision support system (DSS) as a case study.