CNN FEATURES ARE ALSO GREAT AT UNSUPERVISED CLASSIFICATION

 Author :  Joris Guérin

Affiliation :  Laboratoire des Sciences de l'Information et des Systèmes (CNRS UMR 7296) Arts et Métiers ParisTech

Country :  France

Category :  Computer Science & Information Technology

Volume, Issue, Month, Year :  8, 3, February, 2018

Abstract :

This paper aims at providing insight on the transferability of deep CNN features to unsupervised problems. We study the impact of different pretrained CNN feature extractors on the problem of image set clustering for object classification as well as fine-grained classification. We propose a rather straightforward pipeline combining deep-feature extraction using a CNN pretrained on ImageNet and a classic clustering algorithm to classify sets of images. This approach is compared to state-of-the-art algorithms in image-clustering and provides better results. These results strengthen the belief that supervised training of deep CNN on large datasets, with a large variability of classes, extracts better features than most carefully designed engineering approaches, even for unsupervised tasks. We also validate our approach on a robotic application, consisting in sorting and storing objects smartly based on clustering.

Keyword :  Transfer learning, Image clustering, Robotics application

For More Details :  https://airccj.org/CSCP/vol8/csit88308.pdf

CONCEPTUALIZING AI RISK

 Author :  Martin Ciupa

Affiliation :  1CTO calvIO Inc., Webster, New York

Country :  USA

Category :  Computer Science & Information Technology

Volume, Issue, Month, Year :  8, 3, February, 2018

Abstract :

AI advances represent a great technological opportunity, but also possible perils. This paper undertakes an ethical and systematic evaluation of those risks in a pragmatic analytical form of questions, which we term ‘Conceptual AI Risk analysis’. We then look at a topical case example in an actual industrial setting and apply that methodology in outline. The case involves Deep Learning Black-Boxes and their risk issues in an environment that requires compliance with legal rules and industry best practices. We examine a technological means to attempt to solve the Black-box problem for this case, referred to as “Really Useful Machine Learning” ( RUMLSM ). DARPA has identified such cases as being the “Third Wave of AI.” Conclusions to its efficacy are drawn.

Keyword :  AI Risk, Deep Neural Network, Black-box Problem, Really Useful Machine Learning, RUMLSM, DARPA Third Wave of AI. 

For More Details :  https://airccj.org/CSCP/vol8/csit88307.pdf

EXAMINING RISK COMMUNICATION OF SAMSUNG BY ANALYSING CUSTOMER REACTION ON SOCIAL MEDIA: FROM GALAXY NOTE 7 EXPLOSION TO GALAXY S8 RELEASE

Author :  Jaewon Lim

Affiliation :  Department of Interaction Science, Sungkyunkwan University

Country :  Korea

Category :  Computer Science & Information Technology

Volume, Issue, Month, Year :  8, 3, February, 2018

Abstract :

Crisis of product defect causes recalls, which not only causes financial loss but also devastates brand reputation. This study explores customers’ response to the twice recall of the Galaxy Note 7 in Korea. The authors examined the semantic networks of consumers comments from the Samsung official social media ‘Samsung Newsroom (blog and Facebook)’ in five time periods(i.e., release, first recall, recommendation to discontinue use, second recall, and compensation program). Customers in each period showed different emotional states which were positive in the first recall but negative in the second recall. Therefore, timely recall and factual communication is needed to the product-harm crisis management, but root cause investigation should be taken to avoid further recalls.

Keyword :  Galaxy Note7, Product harm crisis, Recall, Semantic network analysis

For More Details :  https://airccj.org/CSCP/vol8/csit88306.pdf

PERCEPTION OF HOME-SHARING INDUSTRY: A STUDY OF COMPARING AIRBNB AND COUCHSURFING

Author :  JinTao Huang

Affiliation :  Department of Interaction Science, SungKyunKwan University, Seoul

Country :  South Korea

Category :  Computer Science & Information Technology

Volume, Issue, Month, Year :  8, 3, February, 2018

Abstract :

The aim of this paper is to examine people’s perceptions of home sharing industry through comparing Airbnb and Couchsurfing, using python to crawling twitter’s data. We conducted semantic network analysis by using UCINET, which is embedded with NETDRAW, for calculating betweenness centrality and visualizing semantic network based on multidimensional scaling(MDS). And we also used LIWC to analysis public sentimental perceptions. The present study’s results show that the characteristics discovered about Airbnb and Couchsurfing in cyberspace have possible future directions in view of word usage frequency, centrality and semantic networks. Moreover, the results show that in sentimental aspects, there are different public trends of emotions on Airbnb and Couchsurfing. Through those results, researchers provide information to understand which sectors should entrepreneurs put more efforts and money.

Keyword :  Sharing Economy, Home-sharing, Airbnb, Couchsurfing, Semantic Network Analysis

For More Details :  https://airccj.org/CSCP/vol8/csit88304.pdf

ON THE DUALITY FEATURE OF P-CLASS PROBLEMS AND NP COMPLETE PROBLEMS

 Author :  WenhongTian

Affiliation :  1University of Electronic Science and Technology of China, Chengdu

Country :  China

Category :  Computer Science & Information Technology

Volume, Issue, Month, Year :  8, 3, February, 2018

Abstract :

In term of computational complexity, P-class (abbreviated as P) problems are polynomial-time solvable by deterministic Turing machine while NP complete (abbreviated as NPC) problems are polynomial-time solvable by nondeterministic Turing machine. P and NPC problems are regularly treated in different classes. Determining whether or not it is possible to solve NPC problems quickly, is one of the principal unsolved problems in computer science today. In this paper, a new perspective is provided: both P problems and NPC problems have the duality feature in terms of computational complexity of asymptotic efficiency of algorithms.

Keyword :  P Problems; NP problems; NP Complete Problems; the P versus NP Problem; The Duality Feature.

For More Details :  https://airccj.org/CSCP/vol8/csit88303.pdf

PERFORMANCE EVALUATION OF MOBILE IP ON MOBILE AD HOC NETWORKS USING NS2

Author :  Kingsley K. Ofosu

Affiliation :  Department of Computer Science, University of Ghana, Legon-Accra

Country :  Ghana

Category :  Computer Science & Information Technology

Volume, Issue, Month, Year :  8, 3, February, 2018

Abstract :

Mobile computing devices equipped with transceivers form Mobile Ad Hoc Networks (MANET) when two or more of these devices find themselves within transmission range. MANETs are stand-alone (no existing infrastructure needed), autonomous networks that utilise multi-hop communication to reach nodes out of transmitter range. Unlike infrastructure networks e.g. the Internet with fixed topology, MANETs are dynamic. Despite the heterogeneous nature of these two networks, integrating MANETs with the Internet extends the network coverage and adds to the application domain of MANETs. One of the many ways of combining MANETs with the Internet involves using Mobile Internet Protocol (Mobile IP) and a MANET protocol to route packets between the Internet and the MANET via Gateway agents. In this paper, we evaluate the performance of Mobile IP on MANET in Network Simulator 2 (NS2). We have implemented Mobile IP on Ad hoc On-demand Distance Vector (AODV), Ad hoc On-demand Multiple Distance Vector (AOMDV) and Destination-Sequenced Distance Vector (DSDV) routing protocols and compared performances based on Throughput, End-to-End Delay (E2ED), Packet Delivery Ratio (PDR) and Normalized Packet Ratio (NPR). The simulation results suggest that on-demand routing within the MANET better serves Mobile IP on MANETs.

Keyword :  Mobile Ad Hoc Network, Mobile IP, AODV, AOMDV &DSDV

For More Details :  https://airccj.org/CSCP/vol8/csit88302.pdf

LOAD BALANCING MANAGEMENT USING FUZZY LOGIC TO IMPROVE THE REPORT TRANSFER SUCCESS RATE

Author :  Sanghyeok Lim

Affiliation :  College of Information and Communication Engineering, Sungkyunkwan University

Country :  Republic of Korea

Category :  Computer Science & Information Technology

Volume, Issue, Month, Year :  8, 3, February, 2018

Abstract :

A wireless sensor network (WSN) consists of multiple sensor nodes and base stations (BS) that collect information over widely deployed sensor nodes. Sensor nodes have limited energy source and low computing power. Due to those features, there is a disadvantage that user's individual node management is difficult and they are easily captured by attackers. Therefore, efficient energy allocation of nodes is important and network security protocol is needed. The Probabilistic Voting Filtering System (PVFS) is a system that prevents false vote injection attack and false report attack injected from attackers. The reason for the existence of this protocol is for energy management of nodes through defence against those attacks and in order to efficiently manage the network based on PVFS, load balancing of nodes should be performed. In the proposed scheme, fuzzy logic is applied to each cluster head node (CH) to perform load balancing by determine whether to perform a role as a verification node and an event forwarding node. The experiment shows that the event detection rate and the report delivery success rate are improved in proposed scheme compare with original PVFS.

Keyword :  Network Protocols, Wireless Sensor Network, Fuzzy logic, False Report Attack, False Vote Injection Attack

For More Details  :  https://airccj.org/CSCP/vol8/csit88301.pdf

RELIABILITY OF MECHANICAL SYSTEM OF SYSTEMS

 Author :  El Hami Abdelkhalakl

Affiliation :  LMN-INSA Rouen Normandie, Normandy University

Country :  FRANCE

Category :  Computer Science & Information Technology

Volume, Issue, Month, Year :  8, 4, February, 2018

Abstract :

In this paper, we present a new methodology about reliability of systems of systems. We present also an example which combines the information transformation in complex systems and virtual design of this system based on finite element analysis. This example is help to balance the performances and the costs in complex system, or provide the optimal solution in manufacturing design. It can also update the existing design of component by changing the new design of this component.

Keyword :  Reliability, Systems of systems, Simulation

For More Details :  https://airccj.org/CSCP/vol8/csit88410.pdf

DISTRIBUTED SYSTEM APPROACH TO EXPERIMENT REGIONAL COMPETITIVENESS

Author :  Mhamed Itmi

Affiliation :  University of Rouen Normandy

Country :  France

Category :  Computer Science & Information Technology

Volume, Issue, Month, Year :  8, 4, February, 2018

Abstract :

This paper highlights a work under development on a regional competitiveness project. We report on a multi-lateral, multi-scale perspective for building cooperative relationships that enhance competitiveness Regionally. The approach mimics a System of Systems methodology whereby entity relationships are captured and defined along several dimensions involving multiple constituents and multiple domain concerns. We build a serious game that is a distributed business simulator to approach the prototyping of this crossroads between supply chain management, geographical economics and information systems.

Keyword :  Adaptive systems, economic geography, multi agent systems, serious game, system of systems, supply chain management.

For More Details :  https://airccj.org/CSCP/vol8/csit88409.pdf

AN ONTOLOGY-BASED DATA WAREHOUSE FOR THE GRAIN TRADE DOMAIN

Author :  Mhamed Itmi

Affiliation :  LITIS Laboratory, INSA Rouen Normandy, University of Rouen Normandy

Country :  France

Category :  Computer Science & Information Technology

Volume, Issue, Month, Year :  8, 4, February, 2018

Keyword :  Ontology Engineering, Data Warehouse, Sparql Endpoint.

Abstract :

Data warehouse systems provide a great way to centralize and converge all data of an organization in order to facilitating access to the huge amounts of information, analysing and decision making. Actually, the conceptual data-models of data warehouses does not take into account the semantic dimension of information. However, the semantic of data models constitute an important indicator to help users to finds its way in any applications that use the data warehouse. In this study, we will tackle this problem trough using ontologies and semantic web techniques to integrate and model information. The contributions of this paper are an ontology for the field of grain trade and a semantic data warehouse which uses the ontology as a conceptual data-model.

For More Details :  https://airccj.org/CSCP/vol8/csit88408.pdf

A PROPOSED HSV-BASED PSEUDOCOLORING SCHEME FOR ENHANCING MEDICAL IMAGES

Author :  Noura A. Semar

Affiliation :  Menoufia University

Country :  Egypt

Category :  Computer Science & Information Technology

Volume, Issue, Month, Year :  8, 4, February, 2018

Keyword :  Medical imaging, pseudo-coloring, colorization, HSV

Abstract :

Medical imaging is one of the most attractive topics of image processing and understanding research fields due to the similarity between the captured body organs colors. Most medical images come in grayscale with low contrast gray values; which makes it a challenge to discriminate between the region of interest (ROI) and the background (BG) parts. Pseudocoloring is one of the solutions to enhance the visual appeal of medical images, most literature works suggest RGB-base color palettes. In this paper, pseudo-coloring methods of different medical imaging works are investigated and a highly discriminative colorization method is proposed. The proposed colorization method employs HSV/HSI color models to generate the desired color scale. Experiments have been performed on different medical images and different assessment methods have been utilized. The results show that the proposed methodology could clearly discriminate between near grayscale organs especially in case of tumor existence. Comparisons with other literary works were performed and the results are promising.

For More Details :  https://airccj.org/CSCP/vol8/csit88407.pdf

PERTURBED ANONYMIZATION: TWO LEVEL SMART PRIVACY FOR LBS MOBILE USERS

Author :  Ruchika Gupta

Affiliation :  Department of Computer Engineering, S. V. National Institute of Technology, Surat, Gujarat

Country :  India

Category :  Computer Science & Information Technology

Volume, Issue, Month, Year :  8, 4, February, 2018

Abstract :

The use of smart mobile devices like tablets, smart phones and navigational gadgets provide most promising communication and better services to mobile users. Location Based Services (LBS) have become very common in recent years. Mobile users submit their location dependent queries to the untrusted LBS server to acquire a particular service. Ideally, user’s personal information such as location data is supposed to be protected while communicating to LBS and at the same time quality of service must be maintained. Therefore, there is a need to have a balanced trade-off between privacy and quality of service. To fulfil such trade-off, this paper proposes a solution that first forms the cloaking region at mobile device, perform perturbation to handle the problem of trusted third party and the anonymizer further anonymizes the location to remove the problem of enough users required to form the cloaking region. The proposed approach protects the location privacy of the user and also maintains the quality of service by selecting appropriate service to the particular user. The proposed algorithm provides two-level location protection to the user, and thus ensures smart mobility of the LBS user.

Keyword :  Location Privacy, Location Based Services, Perturbation, Anonymity, Point of Interest, Smart Mobility

For More Details :  https://airccj.org/CSCP/vol8/csit88406.pdf

PERFORMANCE ANALYSIS OF SYMMETRIC KEY CIPHERS IN LINEAR AND GRID BASED SENSOR NETWORKS

Author :  Kaushal Shah

Affiliation :  Department of Computer Engineering, S.V.National Institute of Technology

Country :  India

Category :  Computer Science & Information Technology

Volume, Issue, Month, Year :  8, 4, February, 2018

Abstract :

The linear and grid based Wireless Sensor Networks (WSN) are formed by applications where objects being monitored are either placed in linear or grid based form. E.g. monitoring oil, water or gas pipelines; perimeter surveillance; monitoring traffic level of city streets, goods warehouse monitoring. The security of data is a critical issue for all such applications and as the devices used for the monitoring purpose have several resource constraints (bandwidth, storage capacity, battery life); it is significant to have a lightweight security solution. Therefore, we consider symmetric key based solutions proposed in the literature as asymmetric based solutions require more computation, energy and storage of keys. We analyse the symmetric ciphers with respect to the performance parameters: RAM, ROM consumption and number of CPU cycles. We perform this simulation analysis in Contiki Cooja by considering an example scenario on two different motes namely: Sky and Z1. The aim of this analysis is to come up with the best suited symmetric key based cipher for the linear and grid based WSN.

Keyword :  Linear and Grid Based Wireless Sensor Networks, Symmetric Key Based Ciphers, Performance Analysis, Contiki Cooja.

For More Details :  https://airccj.org/CSCP/vol8/csit88405.pdf

STRATEGY OF THE REMOVE AND EASY TBT IN GCC 6 COUNTRIES

Author :  Yong-Jae Kim

Affiliation :  Korea Polytechnic University

Country :  South Korea

Category :  Computer Science & Information Technology

Volume, Issue, Month, Year :  8, 4, February, 2018

Abstract :

The last technical barriers to trade(TBT) between countries are Non-Tariff Barriers(NTBs),meaning all trade barriers are possible other than Tariff Barriers. And the most typical examples are (TBT), which refer to measure Technical Regulation, Standards, Procedure for Conformity Assessment, Test & Certification etc. Therefore, in order to eliminate TBT, WTO has made all membership countries automatically enter into an agreement on TBT. In this study,the elimination strategy of TBT with aid of technical regulations or standards is excluded, and only the conformity assessment shall be considered as the strategic measure of eliminating TBT in GCC(Gulf Cooperation Council) 6 countries. The measure for every membership country to accord with the international standards corresponding to their technical regulations and standards, is only to present TBT related Specific Trade Concern(STC) to WTO. However, each of countries retains its own conformity assessment area, and measures to settle such differences are various as well. Therefore, it is likely required an appropriate level of harmonization in them to carry forward this scheme. KTC(Korea Testing Certification) written MRA with GCC test & certification company in 2015 years. So Korea exporting company can export to GCC goods with attached test & certification documents in Korea. To conclude, it is suggest MRA for the remove and reduce TBT to increase export and import among countries.

Keyword :  FTA, Standards, Conformity Assessment, TBT(Technical barrier to trade), MRA, WTO

For More Details :  http://airccj.org/CSCP/vol8/csit88404.pdf

UNDERSTANDING PEOPLE TITLE PROPERTIES TO IMPROVE INFORMATION EXTRACTION PROCESS

Author :  Saleem Abulei

Affiliation :  MMIS Department, Chicago State University

Country :  USA

Category :  Computer Science & Information Technology

Volume, Issue, Month, Year :  8, 4, February, 2018

Abstract :

In this paper, we introduce a new approach to tackle the process of extracting information about people mentioned in the Arabic text. When a person name is mentioned in the Arabic text usually it is combined with a title, in this paper the focus is on the properties of those titles. We have identified six properties for each title with respect to gender, type, class, status, format, and entity existence. We have studied each property, identified all attributes and values that belong to each one of them and classified them accordingly. Sometimes person title is attached to an entity; we have also identified some properties for these entities and we show how they work in a harmony with person title properties. We use graphs for the implementation, nodes to represents person title, person name, entity and their properties, where edges are used to present inherited properties from parent nodes to child nodes.

Keyword :  People Titles, Title Properties, NLP, Arabic Language

For More Details :  https://airccj.org/CSCP/vol8/csit88403.pdf

GRC-MS: A GENETIC RULE-BASED CLASSIFIER MODEL FOR ANALYSIS OF MASS SPECTRA DATA

 Author :  Sara Al-Osimi

Affiliation :  Department of Computer Science, Shaqra University

Country :  Egypt

Category :  Computer Science & Information Technology

Volume, Issue, Month, Year :  8, 4, February, 2018

Abstract :

Many studies uses different data mining techniques to analyze mass spectrometry data and extract useful knowledge about biomarkers. These Biomarkers allow the medical experts to determine whether an individual has a disease or not. Some of these studies have proposed models that have obtained high accuracy. However, the black-box nature and complexity of the proposed models have posed significant issues. Thus, to address this problem and build an accurate model, we use a genetic algorithm for feature selection along with a rule-based classifier, namely Genetic Rule-Based Classifier algorithm for Mass Spectra data (GRC-MS). According to the literature, rule-based classifiers provide understandable rules, but not accurate. In addition, genetic algorithms have achieved excellent results when used with different classifiers for feature selection. Experiments are conducted on real dataset and the proposed classifier GRC-MS achieves 99.7% accuracy. In addition, the generated rules are more understandable than those of other classifier models.

Keyword :  Mass spectrometry, data mining, biomarkers, rule-based classifier, genetic algorithm.

For More Details :  https://airccj.org/CSCP/vol8/csit88402.pdf

DATA SCIENCE METHODOLOGY FOR CYBERSECURITY PROJECTS

Author :  Farhad Foroughi

Affiliation :  Institute of Computer Science University of Rostock

Country :  Germany

Category :  Computer Science & Information Technology

Volume, Issue, Month, Year :  8, 4, February, 2018

Abstract :

Cybersecurity solutions are traditionally static and signature-based. The traditional solutions along with the use of analytic models, machine learning and big data could be improved by automatically trigger mitigation or provide relevant awareness to control or limit consequences of threats. This kind of intelligent solutions is covered in the context of Data Science for Cybersecurity. Data Science provides a significant role in cybersecurity by utilising the power of data (and big data), high-performance computing and data mining (and machine learning) to protect users against cybercrimes. For this purpose, a successful data science project requires an effective methodology to cover all issues and provide adequate resources. In this paper, we are introducing popular data science methodologies and will compare them in accordance with cybersecurity challenges. A comparison discussion has also delivered to explain methodologies’ strengths and weaknesses in case of cybersecurity projects.

Keyword :  Cybersecurity, Data Science Methodology, Data-Driven Decision-making, User Data Discovery, KDD Process, CRISP-DM, Foundational Methodology for Data Science, Team Data Science Process

For More Details :  https://airccj.org/CSCP/vol8/csit88401.pdf

A COMPARATIVE STUDY FOR ICA MULTIUNIT ALGORITHMS

Author :  Doru CONSTANTIN

Affiliation :  Department of Mathematics-Informatics, University of Pitesti

Country :  Romania

Category :  Computer Science & Information Technology

Volume, Issue, Month, Year :  7, 18, December, 2017

Abstract :

We present the comparative study of convergence for multiunit algorithms based on negentropy function for estimating the independent components.

Keyword :  Independent Component Analysis (ICA),Blind Source Separation (BSS), Signal Processing, Negentropy function

For More Details :  https://airccj.org/CSCP/vol7/csit78001.pdf

SOFTWARE FOR THE APPLICATION OF MULTIFRACTAL ANALYSIS IN THE CLASSIFICATION OF METASTATIC BONE TUMORS

Author :  Aleksandar Stančić

Affiliation :  School of Computing, Union University

Country :  Serbia

Category :  Computer Science & Information Technology

Volume, Issue, Month, Year :  8, 9, June, 2018

Abstract :

Multifractal analysis emerged as an accurate tool for cancer classification, although the process needs complete automation in order to be applied in medical diagnostics. After the multifractal analysis has been performed, the results must be processed and properly classified. This paper presents the automation of the diagnostic process. The goal is to successfully apply multifractal analysis as an auxiliary diagnostic tool.

Keyword :  Diagnostic agent, metastatic tumor, multifractal analysis

For More Details :  https://airccj.org/CSCP/vol8/csit88908.pdf