Recent Advances in Computer Science and Communications - Volume 14, Issue 1, 2021

Finding High Utility Itemsets (HUIs) is one of the major problems in the area of frequent itemsets mining. However, HUIs mine lots of redundant itemsets which degrade the performance and importance of high utility itemsets mining. For overcoming this limitation, closed HUIs mining has been proposed. Closed high utility itemsets mining finds complete and non-redundant itemsets. In this paper, we give recent studies on closed high utility itemsets mining algorithms. The main goal of this survey is to provide recent studies and future research opportunities. We give taxonomy of closed high utility itemsets mining algorithms. This paper provides a rough outline of the recent work and gives a general view of closed high utility itemsets mining field.

Influence Maximization, which selects a set of k users (called seed set) from a social network to maximize the expected number of influenced users (called influence spread), is a key algorithmic problem in social influence analysis. In this paper, we give recent studies on influence maximization algorithms. The main goal of this survey is to provide recent studies and future research opportunities. We give taxonomy of influence maximization algorithms with the comparative theoretical analysis. This paper provides a theoretical analysis of influence maximization problem based on algorithm design perspective and also provides the performance analysis of existing algorithms.

Background: Bat Algorithm is one of the swarm intelligence techniques inspired from the echolocation of bats. In this work, many variants of Bat Algorithm are studied which are developed by various researchers. Despite its drawback of getting trapped in local optima, it is preferred over other swarm intelligence techniques. Considering the performance of Bat Algorithm and to extend the existing work, biological behavior of bats is explored in this research work. Objective: One of the characteristics of real bats, i.e. range determination, was adopted to propose a new variant of Bat Algorithm. Methods: The proposed algorithm computed “distance” using cross correlation of emitted pulse and received echo. Results: The performance of Range Determiner-Bat Algorithm (RD-Bat Algorithm) was compared with Standard Bat Algorithm on the basis of best, median, mean, worst and standard deviation values. Conclusion: Experimental results of proposed algorithm outperformed the standard Bat Algorithm.

Background: Intrusion Detection System (IDS) is a Software application which gives the facility to monitor the traffic of network, event or activities on network and finds out any malicious operation if present. Objective: In this paper, a new protocol was developed that can detect the Wireless Network Attack based on the reference of TCP/IP Model. In the proposed system the new feature is integrated in the IDS which are built in the router itself. Methods: If any intruder tries to connect with router, intruder has to authenticate himself/herself. To find the authentication key the intruder attacks on the router to matches the authentication key with the key which he/she has. The intruder has a file with the multiple different keys in it and with that file intruder applies a brute-force attack on the router, the brute-force checks every key of the file by applying them on the router when a key matches with the authenticated key the brute-force software inform the intruder about the key matching. The IDS of the router will checks the rapid tries arriving from the same MAC address, if any MAC address tries the false key many of time than the IDS will identify the MAC as intruder and inform the system administrator about the intrusion by popping up a message on the system of the administrator. Results: Simulation of the two different scenarios is done by using the Network simulator (NS 2) and NAM (Network animator). In scenario 1 the node 1 is intruder and the IDS protocols have figure it out. The intruder is labeled as 2. In scenario 2 node 1 is the sentinel node and it gets connected to router after authentication. Conclusion: The mechanism can detect a false node in the network which is major threat in WSNs. Result has been evaluated the performance of IDS protocol by using Ad-hoc On Demand Distance Vector (AODV) Routing Protocol for routing.

Objective: The blended fusion of Support Vector Machine (SVM) and Principal Component Analysis (PCA) have been widely used in recognizing handwritten digit characters of Devanagari script. The feature information from the character is extracted using its skeleton structure which optimally reduce data dimensionality using PCA. There is ample information available on handwritten charac-ter recognition on Indian and Non-Indian scripts but very few article emphasized on recognition of Devanagari scripts. Therefore, this paper presents an efficient handwritten Devanagari character recognition system based on block based feature extraction and PCA-SVM classifier. Methods: We have collected samples of handwritten Devanagari characters from different handwritten experts for classification. Results: For experimental work, total of 100 images having Devanagari digit characters been used for the purpose of training and testing. The proposed system achieves a maximum recognition accuracy of 96.6 % and 96.5% for 5 & 10 fold validations with 70% training and 30% testing data using block based feature and SVM classifier having different kernels. Conclusion: The obtained results achieve maximum accuracy using SVM classifier for digit character recognition. In future deep learning networks will be considered for accuracy enhancement and precision.

Background: The jammer in a wireless sensor network is located and tracked with open access and shared nature of the wireless medium. The existing algorithms mainly track the stationary jammer. Mobile jammer often moves from one place to another becoming difficult. Mobile jammer location tracker algorithm is proposed to find the location of a mobile jammer with four steps selection i.e., initial examining node, determination of supporting node, trilateration localization and examining group handover. Objective: In this research paper an algorithm is proposed for finding the location of mobile jammer in wireless sensor network. Finding location faces a huge difficulty due to non-supportive working between the multi hop wireless sensor network and jammer. The existing algorithms are used to find the location of stationary jammers only. Mobile jammers frequently change their position from time to time. Therefore jamming increases between the node to node communications in the multi hop wireless sensor network. Methods: The multi hop wireless sensor network is deployed with n number of stationary nodes in a particular area. The omni directional antenna is fitted with those stationary nodes and the transmission powers of the nodes are also constant. The direction-finding table is maintained and keep on updating for an interval of each node. The position of the nodes is tracked with the GPS devices or by existing location finding algorithm. The nodes in the multi hop wireless sensor network are installed in A*A square area uniformly and randomly. The transmitting node is tp node. The received signal to noise ratio is threshold STNR. Neighbor list records the neighbor node of the multi hop wireless sensor networks. Results and Conclusion: The proposed idea of localization and tracking system of mobile jammer is estimated efficiently by the multi hop wireless sensor network with the simulator MATLAB. The nodes in multi hop wireless sensor network are installed in A*A area square uniformly and randomly.

Background: Early detection of cervical cancer may give life to women all over the world. Pap-smear test and Human papillomavirus test are techniques used for the detection and prevention of cervical cancer. Objective: In this paper, pap-smear images are analysed and cells are classified using stacked autoencoder based deep neural network. Pap-smear cells are classified into 2 classes and 4 classes. Twoclass classification includes classification of cells in normal and abnormal cells while four-class classification includes classification of cells in normal cells , mild dysplastic cells, moderate dysplastic cells and severe dysplastic cells. Methods: The features are extracted by deep neural networks based on their architecture. Proposed deep neural networks consist of three stacked auto encoders with hidden sizes 512, 256 and 128, respectively. Softmax used as the outer layer for the classification of pap smear cells. Results: Average accuracy achieved for 2-class classification among normal and abnormal cells is 98.2 % while for 4-class classification among normal, mild, moderate and severe dysplastic cells is 93.8 % respectively. Conclusion: The proposed approach avoids image segmentation and feature extraction applied by previous works. This study highlights deep learning as an important tool for cells classification of pap-smear images. The accuracy of the proposed method may vary with the different combination of hidden size and number of autoencoders.

Background: Extreme growth of data necessitates the need for high-performance computing. MapReduce is among the most sought-after platform for processing large-scale data. Research work and analysis of the existing system has revealed its performance bottlenecks and areas of concern. MapReduce has the problem of skew on its processing nodes. This paper proposes an algorithm for MapReduce to balance the load and eliminate the skew on Map tasks. It reduces the execution time of job by lowering the completion time of the slowest task. Methods: The proposed method performs one-time settlement of load balancing among the Map tasks by analyzing the expected completion time of the Map tasks and redistributes the load. It uses intervals to migrate the overloaded or slows tasks and append them on the under loaded tasks. Results: Experiments revealed an improvement of up to 1.3x by implementing the proposed strategy. Comparison of the proposed technique with other relevant strategies exhibits a better distribution of load among Map tasks and lower level of the skew. Evaluation is done using different workloads. Conclusion: A significant improvement is observed in the performance and reduced completion time of job.

Background: Despite so many constraints, the limited battery power of the sensor nodes is the core issue in Wireless Sensor Networks. This compels how to extend the lifetime of the network as long as possible. One of the ways to solve the problem is to balance the relay traffic load to extend the lifetime. Objective: In this paper, a load balancing algorithm is suggested that selects the best possible relay node so that uniform consumption of the battery power of the sensor nodes can be ensured. Methods: After random deployment, sensor nodes collect information about their neighbors and their expected load. The selection of new next hop starts from maximum hop count. Next hop of the nodes having a single parent is set first. Remaining nodes select their next hop in the non-increasing order of their load. Results: Simulation results verify that packet delivery ratio for proposed work is up to 50% till 72% of total time duration and no nodes getting dead till 48% of total time duration, while for others, nodes start getting dead around 36% of total time duration. Also, it is proved that the solution obtained by proposed work can be at most 1.5 times imbalanced as compared to the optimal solution which implies our solution is quite near to the optimal one. Conclusion: Load balancing done in our work has shown more positive results in comparison to others in terms of network lifetime and first node death and which is also verified with F-test with α- value to be 0.05.

Background: Computational offloading is emerging as a popular field in Mobile Cloud Computing (MCC). Modern applications are power and compute-intensive which leads to the energy, storage and processing issues in mobile devices. Using the offloading concept, a mobile device can offload its computation to the cloud servers and receives back the results on the device. Objective: The main objective of the work is to provide a solution of an important question that arises in the offloading scenario is that which part of the application needs to be offloaded remotely and which part would run locally. Methods: In order to identify remote and local code, the application needs to be partitioned. In this paper, the graph partitioning approach is considered which is based upon the spectral graph partitioning with the Kernighan Lin algorithm. An application is assumed to be a graph and each node of the graph is assumed as a method. Results: Experimental results show that the proposed hybrid approach performs optimally in partitioning the application. The results indicate that considering the combination of spectral approach with the Kernighan Lin algorithm performs optimally as compared to random and multilevel partitioning in a mobile cloud scenario. Conclusion: The proposed technique gave better results than the existing techniques in terms of edge cut which is less, concluding minimum communication cost among components and thus save energy of the mobile device.

Background: Nowadays, the demand for data science-related job positions have seen a huge increase due to the recent data explosion incurred by the industries and organizations globally. The necessity to harness and utilize the amount of information hidden inside these huge datasets for effective decision-making has become the need of the hour. However, this scenario is where a data analyst or a data scientist comes into play. They are domain experts who have the skillset and expertise to extract hidden meaning from data and convert them into useful insights. This work illustrates the use of data mining and advanced data analysis techniques such as data aggregation, summarization along with data visualization using R tool to understand and analyse the job trends in the United States of America (USA) and then drill down to analyse job trends for data science-related job positions from year 2011 to 2016. Objective: This paper discusses the general job trends in the US and how the job seekers are migrating from one place to another place using Visa for different titles, majorly for business analytics. Methods: Analytics is done using R programming, different functions of the programming on various parameters and inference is drawn on the result. Results & Conclusion: The aim of this analysis is to predict the job trends in line with demand, region, employers, wages in USD.

Background: In WDM networks, there is a crucial need to monitor signal degradation factors in order to maintain the quality of transmission. This is more critical in dynamic optical networks as non-linear impairments are network state dependent. Moreover, PLIs are accumulative in nature, so the overall impact is increased tremendously as the length of signal path is increased. The interactions between different impairments along the path also influence their overall impact. Objective: Among the different impairments, the present work focuses on phase modulations owing to the intensities of signals themselves as well as the neighboring signals. It includes the influence of SPM, SPM and XPM, system parameters like signal power, wavelength and fiber parameters like attenuation coefficient, dispersion coefficient and their influence on Q-value and BER. Methods: The analysis is done through a single and two-channel transmitter system with varied power, wavelengths and system parameters. The corresponding optical spectrums are analysed. Results and Conclusion: It has been found that SPM and XPM pose broadening effect on spectrum without any effect on temporal distributions. The magnitude of signal power is among the parameters significantly influencing the broadening of spectrum. The higher the power, the more the magnitude of broadening. It has been found that in order to neglect the impact of input power; its magnitude must be kept below 20 mW. Also, the dispersion and attenuation value need to be carefully as they pose counteracting effect to SPM and XPM for certain values and hence can be used as compensation measures without any additional cost.

Background: The invention of WDM technology in optical communication system has completely revolutionized the telecomm industry through its high data carrying capacity and efficiency of transmission. Advanced optical modulation formats with high spectral efficiency, advanced components like Reconfigurable Optical Add Drop Multiplexers (ROADMS), OXC, and large bandwidth requirements contributed significantly in existence of dynamic, flexible translucent and transparent networks. In these networks, it is common practice to increase the power levels as much as possible to overcome the power penalty effects and better transmission, but this introduces several non-linear impairments in the link and hence degrades the quality of signal flowing. These impairments arise when several high strength optical fields of different wavelengths interact with molecular vibrations and acoustic waves. The different non-linear impacts include Self Phase Modulation (SPM), Cross Phase Modulation (XPM), Four Wave Mixing (FWM) and scattering effects like Stimulated Raman Scattering (SRS), Stimulated Brillouin Scattering (SBS). The main cause of these impairments is variation in refractive index of fiber (also called Kerr effect) due to intensity of signal flowing through fiber. Due to the degradation impact posed by these impairments, it is crucial to analyze their cause, their influence on system performance and mitigation techniques so as to improve the overall quality of transmission. The monitoring of impairments is quite a challenging task due to their dependency on time, present state of network, signals flowing in adjoining channels and fibers. Objective: The present work aims to identify and describe the role of FWM in optical networks. The mathematical model of FWM is studied to know the parameters influencing the overall impact on system performance. The power of optical source, channel spacing, distance of transmission and presence of dispersion are considered as key factors influencing FWM power being developed. Their impact on FWM power and hence, FWM efficiency is calculated. In addition, the influence of FWM on Quality of transmission is quantified in terms of BER and Q-factor. Methods: The analysis is done through a two-channel transmitter system with varied power, channel spacing, distance of transmission and presence of other degradation factors (dispersion) is taken into account. The corresponding optical spectrums are analysed. Result: In this paper, the non-linear impairment FWM posing degradation effect on the signal quality has been discussed. The basics involved are presented along with the mathematical model. It has been found that FWM results in power transfer from one channel to generation of new waves which may lead to power depletion and interference. The new waves generated depend on the number of wavelengths travelling in the fiber and channel spacing. The influence of FWM on system performance is presented in terms of BER and Q-value. Conclusion: It has been concluded that the increased power of transmission and decreased channel spacing are the crucial factors increasing the magnitude of FWM and need to be closely monitored. On the other hand, increased distance of propagation and presence of certain level of dispersion leads to decrease in FWM power. Therefore, if selected carefully, they may act as source of FWM mitigation without requiring any external compensating device.

Background: Fuel is the most important source of energy essential for humanity. It plays a vital role in the economy of a country. Dependency on non-renewable resources like petroleum brings about manipulation of the oil market and an increase in the price of these resources. Excessive use of non-renewable resources leads to global warming. These resources exist in a finite quantity in nature, which makes renewable resources appealing. Objective: Petroleum being a non-renewable energy mode requires monitoring and dynamic pricing based upon the consumption to reduce excessive usage. This dynamic price control on the basis of consumption developed upon the rules of Novel algorithm for resource allocation in the system will orient the end user to move towards developing alternate resources for power generation in the field of renewable energy like wind, hydro, solar, etc. This orientation is one of the key steps for developing a smart city with efficient resource allocation as the key factor of development. Methods: In order to manage resources efficiently, the following processes are incorporated. Provide End-user a unique ID card (Named as Tra-Card). Develop data of usage (Fuel, Health Resources, and Supply Chain Management) for initial days. Automated Management of Resources using Novel Framework Algorithm with fuzzy based stable marriage algorithm). Develop Customer Management Policy as per the criterion set by governments. Generate Data and rules for use at individual points of work. Results: The process will help the government make efficient policies and produce real time deliverables over complete control and handling of fuel management; the most important of human life need. This will also allow government’s to provide a solution for end to end charging by making fuel an item of luxury and taxing the people who use it as a luxury product. Discussion: Fuel being primary requisite for human survival needs strict monitoring for irregular usage and laundering. As, the fuel has become an item of luxury for few and item for survival of few “it must be taxed as per the usage” of end-user. The Data is experimental and developed using a multiplication factor α which is multiplied to the base fuel price once it crosses a certain range of Kilometer travelled by the user with his car. Conclusion: The model proposed in the paper captures the necessity of development of an efficient method that considers the finiteness of fossil fuels by monitoring the distribution of fuel and its consumption. The purpose of this project is to save energy with aim of AI Engine managed logistics and goal of creating Energy-Efficient survival of the human species.

Aims: Internet of Things (IoT) is the evolution of the Internet designed to sense, collect, analyze and distribute the data via IoT devices that form its core component. An important aspect of pervasive IoT applications is its resource-constrained devices. Most of the real-time Edge-IoT applications generate a huge amount of data, which add to the resource consumption in these devices. To save resources in such applications, efficient node deployment and data aggregation techniques can be used. This paper presents the design and modeling of node deployment and data aggregation in Edge-IoT applications along with the homogeneous and heterogeneous network scenarios for smart agriculture. Objectives: To save resources in such applications, efficient node deployment and data aggregation techniques can be used. This paper presents the design and modeling of node deployment and data aggregation in Edge-IoT applications along with the homogeneous and heterogeneous network scenarios for smart agriculture. Methods: For heterogeneous scenarios, we propose a clustering approach, Superior Aggregator Resource Efficient Clustering (SAREC), to address the resource constraints in pervasive Edge-IoT applications. The comparison of homogeneous and heterogeneous networks is based on LEACH and SAREC protocols, respectively. Results: The results show that SAREC is 25% more efficient in energy utilization and network lifetime than LEACH. The results also show that SAREC is more efficient in terms of storage and processing time as compared to LEACH. Conclusion: Node deployment is an important aspect in determining the architecture, which plays an important role in resource management in pervasive applications of IoT. The IoT nodes are distributed in a selected geographical location and the topology of the network is pre-decided to form an Edge-IoT network. In such an environment, the nodes are deployed to sense, aggregate and analyze the data. This paper presents a pervasive Edge-IoT network along with the mathematical modeling consisting of deployment and aggregation models. An Edge-IoT network for smart agriculture has been deployed and analysis of resource utilization has been performed in homogeneous and heterogeneous scenarios of the network. The resource limitations in pervasive IoT network motivated us to develop a SAREC approach for such Edge-IoT applications that optimizes the use of resources. The comparison of the proposed SAREC protocol is made with respect to LEACH protocol on the basis of energy, network lifetime, number of alive nodes, storage and processing time. The results show that SAREC protocol is 25% more efficient in energy utilization and network lifetime than LEACH. It is also evident from the results that the SAREC is more efficient in terms of storage and processing time as compared to LEACH.

Introduction: Over the past decade Heart and diabetes disease prediction are major research works in the past decade. For prediction of the Heart and Diabetes diseases, a model using an approach based on rough sets for reducing the attributes and for classification, fuzzy logic system is proposed in this paper. Methods: The overall process of prediction is split into two main steps, 1) Using rough set theory and hybrid firefly and BAT algorithms, feature reduction is done 2) Fuzzy logic system classifies the disease datasets. Reduction of attributes is carried out by rough sets and Hybrid BAT and Firefly optimization algorithm. Results & Discussion: Then the classification of datasets is carried out by the fuzzy system which is based on the membership function and fuzzy rules. The experimentation is performed on several heart disease datasets available in UCI Machine learning repository like datasets of Hungarian, Cleveland, and Switzerland and diabetes dataset collected from a hospital in India. The experimentation results show that the proposed prediction algorithm outperforms existing approaches by achieving better accuracy, specificity, and sensitivity.