Recommender Systems are receiving substantial attention in several application areas (such as healthcare systems and e-commerce), where each area has different requirements. These systems are multifaceted by nature. So, many metrics, which are sometimes contradictious, are introduced to assess different aspects. The existence of several alternatives and dimensions to recommendation approaches complicate the evaluation of recommender systems. In such a situation, it is desirable to evaluate and compare recommenders in a united way that assesses the multifaceted aspects of these systems fairly and uniformly. Despite the abundance of evaluation dimensions, the literature still lacks an evaluation method that evaluates the multiple properties of these systems, all at once. As a potential solution, this paper proposes an evaluation methodology that provides a multidimensional assessment of recommender systems. The proposed method, which we call ComPer, combines the most common evaluation dimensions into a single, yet, general evaluation metric. ComPer is inspired by the idea that a recommender system mimics human beings; hence, it can be seen as a human and its outputs can be assessed as human’s outputs. Up to our knowledge, this is the first evaluation approach that deals with recommenders as humans. ComPer aims to be thorough (by combining multiple dimensions), simple (by presenting the final result as a single value), and independent (by providing setting-independent results). The applicability of the proposed methodology is evaluated empirically using three different datasets. The initial results are promising in the sense that ComPer is able to give comparable results regardless of the experimental settings.

With the advent of online social media, such as articles, websites, blogs, messages, posts, news channels, and by and large web content has drastically changed the way individuals take a glimpse at different things around them. Today, it's an everyday practice for some individuals to read the news on the web. Sentiment analysis (also called opinion mining) alludes to the utilization of natural language processing, content investigation, and computational linguistics to distinguish and separate subjective data in source materials. Sentiment analysis is broadly applied to online reviews, news feeds and social networking for a wide variety of applications, ranging from marketing to client services. Sentiment analysis emphasizes on the classification of textual data into positive, negative and neutral categories. This research is an endeavor to the case study that calculates news polarity or emotions on different sports feeds which may influence changes in sports news development patterns. The interest of this approach is to generate various text analytics that computes feelings from all pertinent ongoing sports news accessible out in the public domain. The significance and application value of sentiment analysis of RSS feeds in this study is to distinguish between positive feeds and negative feeds on sports that could affect readers or users minds in order to improve RSS feeds messaging broadcast among folks. The methodology utilizes the sentiment analysis techniques using two different online open-source sentiment analysis tools in Rich Site Summary (RSS) news feeds that have an influence on sports-related broadcast esteems.

In this paper, Analysis and comparison of two popular security verification tools namely Automated Validation of Internet Security Protocols and Applications (AVISPA) and Burrows-Abadi-Needham (BAN) logic are presented in terms of the usability, complexity, and other properties of the selected tools. The comparison shows the benefits and the drawbacks for the two tools. As a case study, two previously proposed security protocols, which were tested before by BAN logic only are evaluated and proved using the automated verification tool AVISPA to ensure that these protocols satisfy the other main security measures.

To discover computing resources available for any application before they are allocated to requests dynamically on demand, developing effective mechanism for resource discovery in Vehicular Cloud Networks (VCN) is very important. Providing the services to the requested vehicle in time is a major concern in the VCN environment. Dynamic and intelligent resource discovery schemes are essential in VCN environment so that services are provided to the vehicles in time. Resource discovery is key characteristic of VCN. VCN requires intelligent algorithms for resource discovery. Creating a mechanism for resource management and search resources is the largest challenge in VCN. There is a need to consider for dynamic way to discover the resources in the VCN. The lack of intelligence in resource handling, less flexible for dynamic simultaneous requests, and low scalability are issues to be addressed for the resource discovery in VCN. In this paper we proposed dynamic resource discovery scheme in VCN. Proposed resource discovery scheme uses Honey Bee Optimization (HBO) technique integrated with static and mobile agents. Mobile agent collects the vehicular cloud information and static agent intelligently identifies the required resources by the vehicle. Dynamic discovery model will take into account different parameters influencing the task execution time to optimize subsequent schedule. To test the performance effectiveness of the scheme, proposed dynamic resource discovery scheme is compared with fixed time scheduling algorithm. The objective of the proposed scheme is to search the resources in VCN with a minimum delay.  The simulation results of the proposed scheme is better than the existing scheme.

This paper focuses on an empirical hand gesture recognition system in the domain of image processing and machine learning. The hand gesture is probably the most intuitive and frequently used mode of nonverbal communication in human society. The paper analyzes the efficiency of the Histogram of Oriented Gradients (HOG) as the feature descriptor and Support Vector Machine (SVM) as the classification model in case of gesture recognition. There are three stages of the recognition procedure namely image binarization, feature extraction, and classification. The findings of the paper show that the model classifies hand gestures for the given dataset with satisfactory efficiency. The outcome of this work can be further utilized in practical fields of real-world applications dealing with non-verbal communication.