Journal of Computer and Robotics
Volume:15 Issue: 2, Summer and Autumn 2022

Requirement engineering is one of the critical phases in the software development process. Functional Requirements (FR) and Non-Functional Requirements(NFR) are two of the fundamental requirements in software projects that are observed in the classifications of most researchers in the software engineering field. Conflicting and overlapping among the requirements in both intra and extra communications levels are one of the main challenges in the elicitation and prioritization phases. This paper presents a decision strategy to respond to this challenge called requirements conflicts management strategy (RCMS). This strategy is defined to manage conflict and overlap of NFRs in the prioritization of the constraints satisfaction model for requirements prioritization, known as "CSOP + RP" model, to which the necessary constraints are applied. RCMS is applied to the "CSOP+RP" model as a pre-processing phase by the requirement analyzer and the results are delivered to the system manager. RCMS is founded on several components: the conflicts catalog among NFRs, the mapping model of NFRs to the domain of software systems, and the calculation of Pearson correlation coefficients in NFRs. The negative, positive, and zero values of the correlation coefficients are calculated on the importance of the requirements, which mean conflict, overlap and neutral, respectively. RCMS was implemented on Police Command-and-Control System(PCCS) as a designed case study with specific NFRs and FRs. Therefore, the statistical analysis of the experimental results shows that the proposed strategy increases the accuracy of the input values of the prioritization model and better decision-making in managing conflicts and controlling overlaps. Furthermore, RCMS help to reduce the ambiguities between NFRs and FRs and also influences of NFRs in requirement ranking by the search-based prioritization approach.

Simultaneous localization and mapping (SLAM) technique is a practical approach for unmanned aerial vehicles (UAVs) to position themselves in unknown zones. In a structured arena with sufficient landmarks and enough lighting, the performance of the existing algorithms is satisfactory. But in a typical indoor field and in absence of GPS signal and poor texture and insufficient lighting, the SLAM would be unstable for navigation owing to the lack of features. In this article's suggested technique, the accuracy and resilience in many unknown situations (including dynamic and static ones) are enhanced by extracting edge and corner features instead of lone point features. A pre-processing block is intended to improve picture frames captured by the RGB-D sensor put on a robot with subpar characteristics. Using a predefined distance function, we filter out dynamic characteristics and solve dynamic issues in the same manner as static problems. Real-time use of our suggested strategy effectively reduces the influence of outliers and moving objects on the SLAM. This improves the accuracy of the procedure's computing output significantly. We validated our findings using data from the Technical University of Munich (TUM) to evaluate the proposed method. Additionally, our developed UAV is utilized for testing as well. The results of the trials indicate that the suggested approach is more precise and less susceptible to changes and system noise than the existing methods.

Graph and hypergraph matching are fundamental problems in pattern analysis problems. They are applied to various tasks requiring 2D and 3D feature matching, such as image alignment, 3D reconstruction, and object or action recognition. Graph pattern analysis considers pairwise constraints that usually encode geometric and appearance associations between local features. On the other hand, hypergraph matching incorporates higher-order relations computed over sets of features, which could capture both geometric and appearance information. Therefore, using higher-order constraints enables matching that is more robust (or even invariant) to changes in scale, non-rigid deformations, and outliers. Many objects or other entities such as gesture recognition and human activities in the spatiotemporal domain can be signified by graphs with local information on nodes and more global information on edges or hyperedges. In this research, and essential review have been done on the unsupervised methods to explore and communicate meta-analytic data and results with a large number of novel graphs proposed quite recently.

nowadays, there is a growing demand for the use of fog computing in applications such as e-health, agriculture, industry, and intelligent transportation management. In fog computing, optimal offloading is of crucial importance due to the limited energy of mobile devices. In this regard, using machine learning methods has recently attracted much attention. This paper presents a reinforcement learning-based approach to motivate users to offload their tasks. We propose a self-organizing algorithm for offloading based on Q-learning theory. Performance evaluation of the proposed method against traditional and state-of-the-art methods shows that it consumes less energy. It also reduces the execution time of tasks and results in less consumption of network resources.

Due to the growing number of articles and books available on the web, it seems necessary to have a system that can extract users' articles and books from the vast amount of information that is increasing day by day. One of the best ways to do this is to use referral systems. In this research, a method is provided to improve the recommender systems in the field of article recommendation to the user. In this research, DBSCAN clustering algorithm is used for data clustering. Then we will optimize our data using the firefly algorithm, then the genetic algorithm is used to predict the data, and finally the recommender system based on participatory filtering provides a list of different articles that can be of interest to the user. Be him. The results of the evaluation of the proposed method indicate that this recommending system has a score of 94% in the accuracy of the system. And in the call section, it obtained a score of 91%, which according to the obtained statistics, it can be said that this system can correctly suggest up to 90% of the user's favorite articles to the user.

Today, one of the widely used fields in artificial intelligence is text mining methods, which due to the expansion of virtual space and the increase in the use of media and social messengers, and on the other hand, the ability of these methods to extract the desired information from a very large volume of Unstructured text files have a special place. for example, one of its applications can be mentioned in spam detection. Nowadays, the presence of spam content in social media is increasing drastically, and therefore spam detection has become critical. Users receive many text messages through social networks. These messages contain malicious links, programs, etc., and it is necessary to identify and control spam texts and emails to improve social media security. There are various techniques for this, among which neural networks have shown more effective results. In this article, an approach based on deep learning using an LSTM neural network and Glove word embedding method is introduced to display text word vectors to detect spam emails. The results of the proposed model have been evaluated using accuracy criteria. This model has shown successful and acceptable performance by achieving 98.39% and 99.49% accuracy on two different data sets.