International Journal of Scientific Research in Network Security and Communication
Not a member yet
272 research outputs found
Sort by
A Secure Light Weight Authentication Protocol for Wireless Sensor Network in Internet of Things
Get PDFWith the advancement of cloud and Internet of Things (IoT) technology, mobile phones, RFID systems and wireless sensor networks can be integrated to form heterogeneous systems to execute smarter applications. However, data exchange between remote cloud and sensor node via internet poses critical security challenges. The major challenge is the authentication and key exchange among the communication agents. In addition, resource constrained devices such as RFID tags, sensors in WSN and IoT integration (WSNIT) would require robust and light weight authentication schemes. To combat these issues, we establish in this paper a first of its kind of a WSN security protocol in IoT, which is light weight and resistant to cryptographic attacks.
 
Diagnosis of Diabetes Using Bee Colony Algorithm and Fuzzy Decision Tree
Get PDFToday, in medical knowledge, collecting a lot of data about different diseases is very important. Medical centers collect this data for various purposes. One of the goals of using this data is to research these data and obtain useful results and patterns in relation to diseases. The large volume of this data and the confusion used to overcome this problem to obtain useful relationships between risk factors in diseases. In this study, due to the importance of diabetes in medicine, the aim is to present a hybrid model using fuzzy decision tree and bee cloning algorithm to increase the accuracy of diagnosis. The proposed method is called ABC-FDT. In ABC-FDT, the number of optimal fuzzy sets for each feature is considered so that the best segmentation for the features is provided with the two goals of accuracy and reduction of complexity. PID diabetes data set and classification methods based on ID3, C4.5 and CART rules were used for evaluation. The results indicate the superiority of ABC-FDT algorithm in terms of rules of number, accuracy, sensitivity and specificity. ABC-FDT also outperformed the recently introduced GAANN, QFAM-GA, and SM-RuleMiner algorithms by 10.1, 5.4, and 7.8 percent, respectively
Providing a Web Recommender System Using Markov Chains and Registration Files Structure
Get PDFToday, due to the increasing growth of the internet and the huge amount of information we need systems to be able to recommend the most appropriate services and products to the user. The systems that do this are recommender systems. These systems are intelligently using artificial intelligence techniques to identify the interests of your users on the internet and suggest tailored offers to the user’s preferences and interests. Today, Markov models commonly used to predict web pages. For this purpose, in this research, we use a new Markov model and use the structure of registration files to predict the next pages obtained by the user. The proposed Markov model is based on a matrix of 1 to k and in the form of a Markov model which predicts the next pages. In order to reduce the complexity of the search space, as well as to better navigate to the recommender system, we use k-means clustering to group users. The results of the evaluations of the proposed method on the NASA web server log file and in the F-Measure criterion shows 0.57% superiority over BCF.
 
An Audit of Indian Clinical Laboratories’ websites: Analysis of Optimization Suitability for Search Engines and Digital Marketing
Get PDFSearch engine optimization is an important aspect of the digital world we live in today. An extensive literature review showed only an occasional report of analysis of search engine optimization of medical services’ websites. We analyzed 16 websites of clinical laboratories in India for their search engine optimization status through features like secured socket layer, robots.txt and sitemap.xml file, page quality-related tags, page loading speed, link analysis, structured data, and “viewport” tag for mobile responsiveness. Few shortcomings were identified such as thin and duplicate page content in 50% of the websites, lack of structured data in 44.75%, and absence of robots.txt, sitemap.xml files and ‘alt’ and ‘srcset’ tags from 18.75%, 25%, 56.25% and 87.5% websites, respectively. Avoiding these short comings may improve organic traffic to the websites beneficial to both the marketer and the consumers.
 
An efficient and secure ID based Asymmetric Key Management Scheme for Dynamic Groups
Get PDFA Group Key Agreement (GKA) protocol in Mobile Adhoc Networks (MANETS)permits a group of users to share a common secret using wireless links. Due to the advancement of technology, GKA poses a major challenge to provide authentication & confidentiality in open networks. This paper provides the diversified details of distributed group key management protocols from centralized group key management protocols and traditional symmetric group key agreement from asymmetric group key agreement protocols. In asymmetric group key management protocols, only a group encryption key is negotiated, and decryption key is kept secret at each user. In our proposed ID-based Asymmetric distributed group key agreement protocol, Chinese Remainder Theorem is used for computing Group Encryption Key and RSA algorithm is used for generating public and private keys for each member. The main advantage of our protocol is that it decreases the computation overhead of the members with respect to group dynamics and provides secure multicast communication in MANETs in comparison with other existing schemes.
 
Multi-Level and Mutual Log Integrity Preservation Approach for Cloud Forensics using Public Key Infrastructure
Get PDFThe increasing growth of cloud computing makes the forensic investigation complex. It is very significant to collect and preserve the admissible evidences of different actions happened in the cloud environment in spite of collusion problem. The forensic investigation of logs poses a great challenge to make sure that the investigated logs are not tampered and consistent. Various integrity preservation methods are developed to secure the trustworthy evidences of cloud, but storing the logs of user actions and to ensure the trustworthiness with respect to forensic investigation still result a challenging issue in cloud. Hence, an effective multi-level and mutual log integrity preservation method is developed to assure the integrity of log evidence using public key infrastructure. The integrity preservation mechanism is modelled by employing the security mechanisms, such as hashing, encryption, decryption, and walsh transform. The communication between the entities is achieved by generating the keys using the hashing and encryption function with the concatenation operator. However, the public key and the random number is used to generate the message that can be exchanged between the cloud entities. The investigation process is done by the investigator for transferring the encrypted log file to the server. The proposed method achieved higher performance by considering the 500 number of cloud users in terms of detection rate, memory, and time with the values of 0.9828, 2185 bytes, and 7287 ms, respectively.
 
Optimizing and Enhancing Performance Classification Algorithm on Heart Disease through Feature Selection
Get PDFThe ever-increasing size of datasets in the Big Data era requires effective methods for extracting meaningful information. Data Mining provides a means to analyze large datasets and uncover valuable patterns that can inform future decisions. In this study, we analyze a healthcare dataset of heart diseases to predict the likelihood of a patient having a heart disease based on specific parameters. To accomplish this, we implement decision tree classification algorithms such as ADTree, J48, and RandomForest. Additionally, a feature selection algorithm is applied to remove the least significant three attributes from the dataset, resulting in improved classification performance. Comparing the previous and current results reveals the effectiveness of this approach in enhancing the classification accuracy.
 
Design of an IoT-based management and monitoring system for intelligent irrigation
Get PDFAgriculture plays a major role in the world economy and most people depend on it for their livelihood. This makes water an important resource that must be conserved using the latest available technologies. Today, the Internet of Things has expanded its capabilities to smart agriculture. In this research, an automated and low-cost system for intelligent irrigation based on a fuzzy-based energy-aware routing approach is presented. In addition, a neural network is trained to determine the best irrigation program, based on information received from sensors (such as temperature, soil moisture, etc.). The user in the system can monitor the data collection process with mobile phones, mobile computers, etc. and manage the irrigation of agricultural products. The proposed system proves its intelligence, low cost and portability, its suitability for greenhouses, farms, etc. The simulation results show that the proposed method offers better results compared to the LEACH protocol as well as the WSN-IoT algorithm in various criteria such as grid life and power consumption
 
Optimization of Sentiment Analysis Methods via Machine Learning Algorithms
Get PDFThis research work focus on the latest studies that have used Machine learning to find a solution of sentiment analysis problems related to sentiment polarization. In preprocessing steps, the Models applied stop words and Bag of Words to a collection of datasets. Even with the widespread usage and acceptance of some approaches, a superior technique for categorizing the polarization of a text documents is tough to make out. Machine learning has lately evoked the attention as a method for sentiment investigation. The present work proposes a machine learning based hybrid algorithm that incorporate N-gram technique as a feature extraction and combines Decision tree classifier and Random forest Classifier techniques as a classification for sentiment analysis. Naïve bayse, linear classifier and support vector machine approaches are perform in the perspective of sentiment classification. Finally, a comparative study with the different supervised algorithm is implemented on product reviews dataset. The performance of models are evaluated on confusion matrix. In the comparative analysis of classification techniques, the combined technique has shown better results than previously used supervised techniques of naïve bayse ,linear classifier and support vector machine.
 
Provide a Dynamic Routing Algorithm for MPLS Networks Using Fuzzy Filtering Approach
Get PDFThe advent of Multiprotocol Label Switching (MPLS) technology is the basis of the next generation network to improve multimedia applications. One of the most basic traffic engineering concepts in MPLS is routing for Layered Service Provider (LSP). The purpose of providing routing algorithms is to maximize the number of routed requests according to QoS satisfaction. Most research in this area focuses solely on bandwidth, and relatively few studies consider both bandwidth limits and delay. In this paper, a dynamic routing algorithm based on a fuzzy filtering method is proposed which considers both bandwidth limits and end to end delay for route search. The fuzzy system is a prediction model based on fuzzy weighting rules for filtering high-resource requests. The proposed method tries to postpone requests with high bandwidth and maximum end to end delay. Various scenarios have been used to simulate and evaluate the efficiency of the proposed method, and criteria such as the number of requests accepted, average path length and load balance have been measured. The simulation results prove that the proposed method provides optimal routing performance for MPLS networks.