Recent Advances in Computer Science and Communications - Volume 14, Issue 4, 2021

Background: Robotic manipulator system has been useful in many areas like chemical industries, automobile, medical fields etc. Therefore, it is essential to implement a controller for controlling the end position of a robotic armeffectively. However, with the increasing non-linearity and the complexities of a robotic manipulator system, a conventional Proportional-Integral-Derivative controller has become ineffective. Nowadays, intelligent techniques like fuzzy logic, neural network and optimization algorithms has emerged as an efficient tool for controlling the highly complex nonlinear functions with uncertain dynamics. Objective: To implement an efficient and robustcontroller using Fuzzy Logic to effectively control the end position of Single link Robotic Manipulator to follow the desired trajectory. Methods: In this paper, a Fuzzy Proportional-Integral-Derivativecontroller is implemented whose parameters are obtainedwith the Spider Monkey Optimization technique taking Integral of Absolute Error as an objective function. Results: Simulated results ofoutput of the plants controlled byFuzzy Proportional-Integral- Derivative controller have been shown in this paper and the superiority of the implemented controller has also been described by comparing itwith the conventional Proportional-Integral-Derivative controller and Genetic Algorithm optimization technique. Conclusion: From results, it is clear that the FuzzyProportional-Integral-Derivativeoptimized with the Spider monkey optimization technique is more accurate, fast and robust as compared to the Proportional- Integral-Derivativecontroller as well as the controllers optimized with the Genetic algorithm techniques.Also, by comparing the integral absolute error values of all the controllers, it has been found that the controller optimized with the Spider Monkey Optimization technique shows 99% better efficacy than the genetic algorithm technique.

Background: Stock price prediction is a challenging task. The social, economic, political, and various other factors cause frequent abrupt changes in the stock price. This article proposes a historical data-based ensemble system to predict the closing stock price with higher accuracy and consistency over the existing stock price prediction systems. Objective: The primary objective of this article is to predict the closing price of a stock for the next trading in more accurate and consistent manner over the existing methods employed for the stock price prediction. Methods: The proposed system combines various machine learning-based prediction models employing Least Absolute Shrinkage and Selection Operator (LASSO) regression regularization technique to enhance the accuracy of stock price prediction system as compared to any one of the base prediction models. Results: The analysis of results for all the eleven stocks (listed under Information Technology sector on the Bombay Stock Exchange, India) reveals that the proposed system performs best (on all defined metrics of the proposed system) for training datasets and test datasets comprising of all the stocks considered in the proposed system. Conclusion: The proposed ensemble model consistently predicts stock price with a high degree of accuracy over the existing methods used for the prediction.

Background: In Energy Harvesting Wireless Sensor Networks (EH-WSNs), sensors are harvesting energy from the renewable environment to make their operations endless and uninterrupted. However, in such a network, the time-varying nature of harvesting imposes a challenging issue in obtaining improved data-throughput. The use of a static-sink in EH-WSNs to improve data- throughput is less reliable because there is no assurance of the network connectivity. To alleviate such shortcomings, a Data Mule (MDM) has been introduced in EH-WSN for collecting sensors’ data. In this article, the MDM-based distance constrained tour finding problem is formulated such that the data-throughput can be improved within a given delay constraint. Methods: To solve the problem, we devise two different heuristic algorithms based on two different metrics. Results: The obtained experimental results demonstrate that the devised algorithms are more effective than the existing algorithms in terms of data-throughput. Conclusion: The data-throughput values of the first proposed algorithm are about 6.14% and 3.56% better than the other for two different data gathering time durations of 100 sec and 800 sec. The data-throughput values of the second proposed algorithm are about 5.03% and 5.25% better than the other for two different data gathering time durations of 100 sec and 800 sec.

Background: Emotional speech synthesis is the process of synthesising emotions in a neutral speech – potentially generated by a text-to-speech system – to make an artificial humanmachine interaction human-like. It typically involves analysis and modification of speech parameters. Existing work on speech synthesis involving modification of prosody parameters does so at sentence, word, and syllable level. However, further fine-grained modification at vowel level has not been explored yet, thereby motivating our work. Objective: To explore prosody parameters at vowel level for emotion synthesis. Methods: Our work modifies prosody features (duration, pitch, and intensity) for emotion synthesis. Specifically, it modifies the duration parameter of vowel-like and pause regions and the pitch and intensity parameters of only vowel-like regions. The modification is gender specific using emotional speech templates stored in a database and done using Pitch Synchronous Overlap and Add (PSOLA) method. Results: Comparison was done with the existing work on prosody modification at sentence, word and syllable label on IITKGP-SEHSC database. Improvements of 8.14%, 13.56%, and 2.80% for emotions angry, happy, and fear respectively were obtained for the relative mean opinion score. This was due to: (1) prosody modification at vowel-level being more fine-grained than sentence, word, or syllable level and (2) prosody patterns not being generated for consonant regions because vocal cords do not vibrate during consonant production. Conclusion: Our proposed work shows that an emotional speech generated using prosody modification at vowel-level is more convincible than prosody modification at sentence, word and syllable level.

Background: Advances in Mobile and Internet technology evolved several online applications like smart class, virtual class and online classes. Online courseware influences better subjective knowledge of the learners. The effectiveness of processes of teaching and learning must evaluated for the benefits of the learners to select the best approach of learning which motivated us to evaluate and compare different Online Learning courses Effectiveness through statistical approaches. Objective: The main objective of this paper is to compare the learning effect of National Program on Technology Enhanced Learning (NPTEL) with traditional class room learning approach. Methods: Master of Science -Final year Computer Science students has been allowed to learn their subjects in online learning mode using with NPTEL and traditional learning approach in two different groups. After learning of the subjects a series of tests has been conducted and their marks are recorded for comparison of two different learning modes For comparison of the results of two learning methodologies two different measuring statistical matrices namely F-test and T-test has been taken. The experimental results demonstrate thatthe t-test results of NPTEL and the f-test results for NPTEL learning method are superior than the other comparative learning methods. Results: The test shows that online learning approach provides better learning as compared to traditional classroom learning. Conclusion: The obtained results also indicate that there is a significant improvement on learners through NPTEL video lectures over traditional class room based learning.

Background: The classification method is required to deduce possible errors and assist the doctors. These methods are used to take suitable decisions in real world applications. It is well known that classification is an efficient, effective and broadly utilized strategy in several applications such as medical diagnosis, etc. The prime objective of this research paper is to achieve an efficient and effective classification method for Diabetes. Methods: The proposed methodology comprises two phases: The first phase deals with t h e description of Pima Indian Diabetes Dataset and Localized Diabetes Dataset, whereas in the second phase, the dataset has been processed through two different approaches. Results: The first approach entails classification through Polynomial Kernel, RBF Kernel, Sigmoid Function Kernel and Linear Kernel SVM on Pima Indian Diabetes Dataset and Localized Diabetes Dataset. In the second approach, PSO has been utilized as a feature reduction method followed by using the same set of classification methods used in the first approach. PSO_Linear Kernel SVM provides the highest accuracy and ROC for both the above mentioned dataset. Conclusion: The present work consists of a comparative analysis of outcomes w.r.t. performance assessment has been done PSO and without PSO for the same set of classification methods. Finally, it has been concluded that PSO selects the relevant features, reduces the expense and computation time while improving the ROC and accuracy. The used methodology could be implemented in other medical diseases.

Background: The voice recognition system is about cognizing the signals, by feature extraction and identification of related parameters. The entire process is referred to as voice analytics. Objective: The paper aims at analyzing and synthesizing the phonetics of voice. The work focuses on the facts of voice analytics i.e. basic blocks of ‘Glottal signature’. The glottal signature and unique voice cues are evaluated to derive the relationship for utterance of emotional words which leads to sentimental expression cues. An effort is made to map further to understand sarcasm behavior in the sounds made by human speech. Methods: The basic blocks of unique features identified in the work are Intensity, Pitch, Formants related to speak, read, interactive and declarative sentences solely on voice cues not on linguistic theory. It is also tested to identify derived features that maps to fine-grained details of voice cues to drill up usage in sarcasm detection. Results: Different unique features identified in the work are, intensity, pitch, formants related to read, speak, interactive and declarative sentences and derived parameters. Conclusion: The work carried out in the paper also supports the analysis of voice segmentation labelling, analyzes the unique features of voice cues, understanding physics of voice, the process is further carried out to recognize sarcasm.

Objective: Ranking in many areas has been a big problem for a long time. The authors tried to use a novel ranking approach to find out the most popular research interest among all the research fields. The authors have assumed that the selection of a research area is a tedious task. Methods: Therefore, they tried to propose a mechanism named METHWORK (Methodology With the Opted Related Keywords) to choose popular research trends. Google-based searching was applied to find samples in the initial state of this approach. The METHWORK was tested with respect to a case of ranking of research trends within the IoET (Internet of Environmental Things). To find ranks, the first phase is to assess the popularity of the topics in the existing published research papers. To find out the correctness of the ranking found using METHWORK, the authors performed χ2 hypothesis testing while making a comparison with current ranking techniques. Results: The methodology proposed is a milestone to find out the research trends within a broad research area. Conclusion: The results of the test indicated that the approach could be applied to determine trends in any research discipline and proved the applicability of the proposed approach.

Background: The process of election prediction started long back when common practice for election predictions were traditional methods like pundits, hereditary factor etc. However, in recent times new methods and techniques are being used for election forecasting like Data mining, Data Science, Big data, and numerous machine learning techniques. By using such computational techniques the whole process of political forecasting is changed and poll predictions are carried out through them. Objective: The main objective of this research work is to propose an election prediction model for developing areas especially for the state of Jammu and Kashmir (India). Methods: The election prediction model is developed in Jupyter notebook web application using different supervised machine learning techniques. To obtain the optimal results, we perform the hyperparameter tuning of all the proposed classifiers. For measuring the performance of poll prediction system we used confusion matrix along with AUROC curve which depicts that this methods can be well suited for political forecasting. An important contribution of this article is to design a Prediction system which can be used for making prediction in other fields like cardiovascular disease predictions, weather forecasting etc. Results: This model is tested and trained with real-time dataset of the state Jammu and Kashmir (India). We applied features selection techniques like Random Forest, Decision Tree Classifier, Gradient boosting Classifier and Extra Gradient Boosting and obtained eight most important parameters like (Central Influence, Religion Followers, Party Wave, Party Abbreviations, Sensitive Areas, Vote Bank, Incumbent Party, and Caste Factor) for poll predictions with their mean weightages. By applying different classifier to get mean weightage of different parameters for this election prediction models, it has been observed that Party wave got maximum mean weightage of 0.82% as compared to others parameters. After obtaining the vital parameters for political forecasting, we applied various machine learning algorithms like Decision tree, Random forest, K-nearest neighbor and support vector machine for the early prediction of elections. Experimental results show that Support Vector Machine outperformed with a higher accuracy of 0.84% in contrast to others classifiers. Conclusion: In this paper, a clear overview of election prediction models, their potentials, techniques, parameters as well as limitations are outlined. We conclude this work by stating that election predictions can indeed be forecasted with significant parameters however, with caution due to the limitations which were outlined in developing nations like sensitive areas, social unrest, religion etc. This research work may be considered as the first attempt to use multiple classifier for forecasting the Assembly election results of the state Jammu and Kashmir (India).

Background: Railway systems all over the world face an uphill task in preventing train delays. Categorically in India, the situation is far worse than other developing countries due to the high number of passengers and poor update of the previous system. As per a report in Times of India (TOI), a daily newspaper, around 25.3 million people used to travel by train in 2006 which drastically increased year on year to 80 million in 2018. Objective: Deploy Machine Learning model to predict the delay in arrival of train(s) in minutes, before starting the journey on a valid date. Methods: In this paper we combined previous train delay data and weather data to predict delay. In the proposed model, we use 4 different machine learning methods (Linear regression, Gradient Boosting Regression, Decision Tree and Random Forest) which have been compared with different settings to find the most accurate method. Results: Linear Regression gives 90.01% accuracy, while Gradient Boosting Regressor measure 91.68% and the most accurate configuration of decision tree give 93.71% accuracy. When the researcher implemented the ensemble method, Random forest regression, the researcher achieved 95.36% accuracy. Conclusion: Trains in India get delayed frequently. This model would assist the Indian railways and concerned companies by giving the possibility of finding frequent delays during certain times of the week. The Indian railways could thereafter implement delay preventions during these particular times of the week in order to maintain a good on-time arrival rate.

Background: Crime against women is a major issue of society, resulted in physical, psychological, sexual or economic harm of women. Objective: The main objective of this research paper is to propose the conceptual model after finding the contextual relationship among all identified factors affecting crime against women. Methods: Interpretive structural modeling (ISM) technique has been used to develop the model. This is the mathematical approach or step-by-step procedure to deploy the unstructured items into a structured or hierarchical form to build a conceptual model. Further, MICMAC (Matriced' Impacts Croisés Multiplication Appliquée á un Classement) analysis has also been applied to segregate the factors into groups of independent, dependent and operational factors on the basis of their driving power and dependence power. Hence, driving power indicates on what extent an individual factor is contributing for driving the issue and dependence power indicates on what extent an individual factor-driven from other factors. Results: As resulted all 11 identified factors have been structured into a well-defined model with the groups of linkage factors, independent factors, and dependent factors. The model clearly defines the role and contribution of each factor which gives very good insights to take any kind of decision by the law firms, police departments and other criminal or crime organizations to control or prevent this major issue ‘crime against women’. Conclusion: Public and private crime organizations, law firms can use this article to reform its policies to take more security measures with their implementation related to it.

Background: Background of this paper has taken place in mainly heterogeneous network in which three types of nodes are present like normal node, advance node and super node with different amount of energy. The energy of super node is greater than that of advance and normal nodes and the energy of advance nodes are also greater than that of normal nodes in the designed network. The optimization techniques have to be studied from the swarm intelligence based on the different aspects of routing. Objective: The objective of this paper is to propose a new heterogeneous protocol with the help of hybrid meta-heuristic technique. In this technique, the shortest route has been selected and forwarded the data to the sink in a minimal time span to save the energy and make the network more stable. Methods: To evaluate the technique, a new hybrid technique has been created, where the data transmission is implemented from the beginning. This technique contains the route process of the algorithm which was made available through a hybrid meta-heuristic technique. Results: Simulation results show that the hybrid meta-heuristic technique has high throughput with less number of dead nodes with existing methods and also show that the efficiency and stability of new proposed protocol. Conclusion: The conclusion to this paper is a novel, energy-efficient technique applied for randomly deployed sensor nodes over the wireless sensor network and enhancement has been done in stability and throughput of a new proposed algorithm in case of static as well as moving nodes.

Background: Wireless Sensor Network (WSN) is a major technology for the Internet of Things (IoT) and is used within an IoT system to facilitate collaboration of heterogeneous information systems and services. Due to its distributed nature, these networks are highly vulnerable to various security threats which adversely affect their performance. Trust is one of the influential factors that applies in the security of WSNs to have its applications in cloud system, e-commerce etc. The secure and efficient neighbor selection is an issue of Multiple Criteria Decision Making (MCDM), where many Quality of Service (QoS) parameters play a vital role in the process of best neighbor selection. Methods: A dynamic and efficient trust model is proposed in this paper based on the ranking method for recommendation of appropriate secure neighbor node. To rank the available neighbors, we use voting approach and also a hybrid method of Analytical Hierarchy Process (AHP) and Technique for Order Preference by Similarity to the Ideal Solution (TOPSIS) methods. Results: The paper included a case study to demonstrate the effectiveness of proposed method which maximizes the defense against internal attacks. Complexity analysis has been done to show the superiority of the proposed method. Time complexity of the proposed algorithm is O (n2) against the compared algorithm the growth rate of which is O (2n). Conclusion: This method evaluates trustworthiness of neighbor node quantitatively as a fraction in the range 0 and 1. The proposed algorithm when applied, selects the best node among the alternatives.