International Journal of Advanced Networking and Applications

4585-4591

Authors: Peter S. Nyakomitta, Dr. Silvance O. Abeka

Data exfiltration is a serious cybercrime facing many organizations worldwide. Over the past few years, notable organizations such as the Google, Yahoo, the Pentagon, Iran nuclear facility and the United States military contractors and banks have fallen victims of data exfiltration. The current techniques for averting these threats revolve around firewalls, intrusion detection systems, intrusion prevention techniques, firewalls, anti-virus an anti-malware. However, despite heavy deployment of these devices, attackers still continue to wreck havoc on organizations and individuals, stealing their sensitive data. The aim of this paper was therefore to explore how the current techniques for data loss prevention fail. The results of this analysis revealed that these techniques either use whitelists, blacklists, signature-based scanning, behavioral analysis of programs which are not sufficient to counter attacks based on zero day vulnerabilities. Based on these shortcomings, a novel data exfiltration prevention algorithm is proposed towards the end of this paper. This algorithm is suggested to employ real-time traffic entropy coupled with heuristically computed functional correlations to detect data exfiltrations. The premises of this algorithm and its operations are discussed at the last section of this paper.

4596-4600

Authors: Sahar Alsafi, Samani A. Talab

Wireless sensor networks(WSN) consists of widespread random deployment of energy constrained sensor nodes, many routing protocols have been proposed based on heterogeneity with main research goals such as achieving the energy efficiency, lifetime, and deployment of nodes. In this paper, we have proposed an energy efficient cluster head scheme, for heterogeneous wireless sensor networks, by modifying the threshold value of a node based on which it decides to be a cluster head or not, called TDEEC (Threshold Distributed Energy Efficient Clustering) protocol. Simulation results show that proposed algorithm TDEEC performs better than DEEC, DDEEC and EDEEC.

4606-4612

Authors: Guo Tao, Wu Shi-qi, Shi Yi-Cheng, Ma Qian-Qian, Tang Zhi-hang

The text topic analysis is the core element of the comprehensive review of clothing products, which can directly understand the views and consumption trends of consumer groups, taking a brand clothing store in JD.com as the research object, by using Python crawler and HANLP natural language processing technology, seven of the top-selling fashion reviews were classified and analyzed. Word frequency statistics, TF-IDF and other methods were used to quantify the text, this paper uses the visualization techniques such as word cloud graph contrast, pyLDAvis dynamic model and Sankey graph to display customers’ attention points and real shopping needs from various angles. The experimental results show that the visual results of online review research based on the theme model of Lda can clearly show the advantages and disadvantages of customer-centered evaluation and clothing, and provide important reference for merchants to improve decision-making and optimize service.