Improvised Collaborative Filtering for recommendation system

Shefali Gupta, Jagannath University, Jaipur, India

ABSTRACT

Collaborative filtering (CF) is one of the most important techniques of recommendation system and has been utilized by many e-commerce businesses to provide recommendation to its users. This paper sheds light on CF and its methods. This paper demonstrates a practical algorithm by leveraging data on user ratings for mobile phone devices and then provides recommendations to the target user based on the ratings given by similar users. It also elaborates an algorithm of CF that overcomes some of the common limitations faced by other algorithms.

KEYWORDS

Recommendation System, Collaborative filtering

The Impact of Test Case Generation Methods on The Software Performance

Dathar Abas Hasan, Duhok Polytechnic University, Iraq – Kurdistan - Duhok

ABSTRACT

The software advancements increase the demands for effective, efficient and complicated software. Due to the huge amounts of software requirements, occurring some errors in certain program partsis possible.This means a challenge for the software producer. An effective test system is necessary for designing reliable programs and avoiding the errors that may appear during the software production. In this review, many techniques are discussed for generating test cases which are a set of conditions thatdetermine whether the designed programs can satisfy the user’s requirements or not.Fuzzy logic utilizes an operational profile in the process of allocating test case to improve the software quality.The design of fault propagation path to predicts the software defects during the test operation, also the automatic generation for PLC test cases that produce a new track through the program codeto minimize the test cases needed for large size program.

KEYWORDS

Software test, test case, program evaluation, reliable software system, test suite.

Examination of the Adoption of Augmented Reality in Al Masjid Al Haram in Makkah Al Mukarramah: a TAM Model Approach

Abdullah F. Basiouni, Associate Professor, Management Sciences Department, Yanbu University College, Yanbu, Saudi Arabia

ABSTRACT

Despite the growing importance and popularity of augmented reality (AR) technologies in hospitality and tourism, little research to identify the degree of user adoption, particularly in the travel sector of Hajj and Umrah, has been conducted. Based on a questionnaire administered to 967 visitors of the Holy Mosque of Makkah, visitor experiences and adoption intention were examined and analyzed through structural equation modeling (SEM). Adoption was explored using the technology acceptance model (TAM). The analysis results revealed a reasonable fit between data collected and the model used: chi2 (1844.91), chi2 / DF (2.5), RMSEA (0.13), CFI (0.75), and all values of Cronbach's alpha are higher than 0.70. AR adoption is affected by its perceived value by Haram visitors to which the benefits of usefulness and ease of use contribute, along with attitude and behavioral intention towards the AR actual use.

Conversational Agent for University Related Faqs: A Usability Study

Víctor Jimenez¹, Oscar Jimenez², Juan Jimenez³ and Juan Carlos Jimenez⁴, ¹Faculty of Engineering, José Carlos Mariátegui University, Moquegua, Peru, ²Faculty of Engineering, Universidad Privada de Tacna, Tacna, Peru, ³Faculty of Engineering, José Carlos Mariátegui University, Moquegua, Peru, ⁴Contracts and services, Southern Peru Copper Corporation, Tacna, Peru

ABSTRACT

The aim of this paper was to assess the usability of a chatbot for frequetly asqued questions of José Carlos Mariátegui University. The study group was 54 students from Administrative Sciences and Strategic Marketing department, who were sent a System Usability Scale (SUS) questionnaire. The results of this study show that Mariateguino Bot (chatbot name), is rated as “Good”, above normal, so the future of chatbots in the academic field is promising.

KEYWORDS

Dialogflow, Chatbot, Natural Language Understanding, Usability, System Usability Scale.

Back-propagation Neural Network-based Method for Predicting the Interval Natural Frequencies of Structures with Uncertain-but-bounded Parameters

Pengbo Wang^1*, Wenting Jiang² and Qinghe Shi³, ¹Institute of Solid Mechanics, Beihang University, Beijing, 100191, China, ²Institute of Engineering Thermophysics, Chinese Academy of Sciences, Beijing, 100190, China and ³School of Materials and Engineering, Jiangsu University of Technology, Changzhou, Jiangsu, 213001, China

ABSTRACT

Uncertain-but-bounded parameters have a significant impact on the natural frequencies of structures, and it is necessary to study their inherent relationship. However, their relationship is generally nonlinear and thus very complicated. Taking advantage of the strong non-linear mapping ability and high computational efficiency of BP neural networks, namely the error back-propagation neural networks, a BP neural network-based method is proposed to predict the interval natural frequencies of structures with uncertain-but-bounded parameters. To demonstrate the proposed method’s feasibility, a numerical example is tested. The lower and upper frequency bounds obtained using the proposed approach are compared with those obtained using the interval-based perturbation method, which is a commonly used method for problems with uncertainties. A Monte Carlo simulation is also conducted because it is always referred to as a reference solution for problems related to uncertainties. It can be observed that as the varying ranges of uncertain parameters become larger, the accuracy of the perturbation method deteriorates remarkably, but the proposed method still maintains a high level of accuracy. This study not only puts forward a novel approach for predicting the interval natural frequencies but also exhibits the broad application prospect of BP neural networks for solving problems with uncertainties.

KEYWORDS

Back-propagation neural network, Natural frequency, Interval parameter, Perturbation method, Monte Carlo simulation

Intrusion Detection based on Graph oriented Big Data Analytics

Ahlem Abid and Farah Jemili, ISITCOM Hammam Sousse Sousse, Tunisia

ABSTRACT

Intrusion detection has been the subject of numerous studies in industry and academia, but cyber security analysts always want greater precision and global threat analysis to secure their systems in cyberspace. To improve intrusion detection system, the visualization of the security events in form of graphs and diagrams is important to improve the accuracy of alerts. In this paper, we propose an approach of an IDS based on cloud computing, big data technique and using a machine learning graph algorithm which can detect in real time different attacks as early as possible. We use the MAWILab intrusion detection dataset . We choose Microsoft Azure as a unified cloud environment to load our dataset on Azure blob storage. We implement the k2 algorithm which is a graphical machine learning algorithm to classify attacks. Our system showed a good performance due to the graphical machine learning algorithm and spark structured streaming engine.

KEYWORDS

Intrusion detection, MAWILab, Apache Spark Streaming, Microsoft Azure Cloud, Graph, Machine Learning, k2 algorithm

Hand Segmentation for Arabic Sign Language Alphabet Recognition

Ouiem Bchir, College of Computer and Information Sciences, Computer Science Department, King Saud University, Riyadh, Saudi Arabia

ABSTRACT

This research aims to separate the hands from the background of colored images representing the Arabic Sign language alphabet gestures. This hand segmentation task is one of the main challenges of image based Sign language recognition systems due to the issue of skin tones variations and the complexity of the background. For this purpose, an efficient system that segment the hand object and separate it from the rest of the image based on deep learning is investigated. More specifically, the DeepLab v3+ network architecture that is a combination of spatial pyramid pooling module and encode-decoder structure will be trained to learn the visual characteristics of the hand and segment it with detailed boundaries. The effectiveness of the proposed solution is investigated on a large dataset of size 12000 with an accuracy of 98%, an IoU of 93% of and BF score of 87%.

KEYWORDS

Hand segmentation, Deep learning, Arabic Sign Language Alphabet, spatial pyramid pooling.

The Study on Economics of Trustworthiness of Blockchain and Smart Contract

Haitao LIU^{1, 2}, Xianzhi WANG³, ¹General Manager of Operation Company of China-ASEAN Information Harbor Co., Ltd., Nanning, 530200, China., ²General Manager of Yulin Data Economy Investment Development Co., Ltd., Yulin, 537005, China and ³DECRA Fellow, School of Computer Science of University of Technology Sydney, NSW, Australia

ABSTRACT

There have an increasing problem that many neonatal families cannot find a trustworthy and suitable postpartum doula when facing with asymmetric information. When they in an effort to find a postpartum doula at offline institutions, it is always inefficient, arduous, expensive, untrustworthiness, unsatisfactory, non evaluation and so on. Therefore, a novel online platform named YBJ was initially developed, which based on smart contract and blockchain, is for conceiving family to precisely find an appropriate postpartum doula efficiently. The YBJ consist of User Contract, Business Contract, Operation Contract. Like the smart contract, these three contracts also will execute the given code automatically. The principle of YBJ was explicitly depicted as well in this paper. As the blockchain, the YBJ also designed having an interesting mining function in additional. And the YBCoin was defined, it was designed as an integrals or virtual currency. It will be rewarded to the legal registered users after mined successfully. In fact, it should be to the benefit of users with this function. In the view of economics and commerce, YBJ and traditional ways was objectively evaluated and compared with some common parameters and index. Of course, as a matter of fact, the online YBJ more obviously favorite for families.

KEYWORDS

blockchain, smart contract, trustworthiness, economics.

Smart Contract based Searchable Symmetric Encryption

Wu Yanbing, Institute for Advanced Study, Tsinghua University, P. R. China

ABSTRACT

Some of the existing Searchable Symmetric Encryption schemes mainly satisfy retrieving requirements from users. After the user has paid, the server will search for the user’s request. Thus there exists such a problem,the server may refuse to the user’s request, reduce search range, or miss some search results in order to save computation cost, bandwidth or some other reasons after payment by the user. For the existing Searchable Symmetric Encryption scheme, users can’t judge for the integrity and accuracy of the results returned by the server. Once the commissions is handed over to the server, even if the result returned by the server is wrong, the user is unable to withdraw the commission. We propose a Searchable Symmetric Encryption scheme called smart contract based SSE. In our scheme, we propose the method to non-interactive verify the integrity of the search results for the first time. We use the Merkle tree to build integrity verification of search results. By adding search result integrity proof, it can ensure that searchers can provide integrity verification without searching all documents. We apply smart contracts, which help us search more efficiently through the computing resources on the blockchain. Neither the owner nor the searcher can cheat, because all information are recorded on the blockchain without tampering. With the addition of integrity verification and smart contracts, our solution does not reduce the efficiency much more and is available in practice. Our scheme ensures that searchers can not get any information from encrypted documents and search keywords. If the server does not search all the documents thoroughly, he will not be able to pass the result integrity test, and will not get any commission. If the search result is verified correctly and is verified through the result integrity, the commission will automatically be transferred from the file owner to the searcher’s account through a smart contract. Our scheme solves the related issues effectively through the proof of result integrity and smart contract, which guarantees the authenticity and the integrity of the results, and the searcher will not be tampering by the miners after the submission of the results through the signature of the Merkle tree root.

KEYWORDS

Smart contract, Blockchain, Searchable Symmetric Encryption, Result integrity, Verifiable, Bloom filter

A Self-attentional Auto Encoder based Intrusion Detection System

Bingzhang Hu and Yu Guan, Open Lab, School of Computing, Newcastle University, Newcastle upon Tyne, United Kingdom

ABSTRACT

Intrusion detection systems (IDSs) have received increasing attention in recent years due to the rapidly development of Internet applications and Internet of Things. Anomaly based IDSs are preferred in many situations due to their capability of detecting novel unseen attacks. However, existing works have neither considered the intrinsic relationships among the network traffic data nor the correlations shared by the sub features (i.e. content feature, host-based feature, etc.) in the network traffic data. In this paper, we propose a self-attentional auto-encoder based intrusion detection system, namely the STAR-IDS and evaluated it on NSL-KDD dataset. The experimental results show that the proposed STAR-IDS have achieved state-of-theart performance.

KEYWORDS

Intrusion Detection System, Auto Encoder, Anomaly Detection, Self Attention, Machine Learning

Decision Rights and Governance within the Blockchain Domain: a Perspective from Practice

Koen Smit, HU University of Applied Sciences, Netherlands

ABSTRACT

In the context of blockchain, governance is one of the fundamental aspects of blockchain design. However, the current body of knowledge on governance, and within that scope, decision rights, is immature in terms of research field maturity. To explore and provide meaning to decision rights in the context of governance, an explorative approach comprising twelve expert interviews and a focus group consisting of five participants. Based on the analysis of the experiences shared, the Blockchain Lifecycle Model (BLM) is created that sheds light on decision rights within the governance of public and private blockchains. The results of this study, together with the contextual BLM model, could structure the discussion and thinking about governance and decision rights in blockchain design. As blockchain pilots and projects evolve to increasingly large communities, governance and decision rights are increasingly important to ensure value-sensitive design for all stakeholders involved.

KEYWORDS

Blockchain, governance, decision rights, expert interviews, focus group

Human Following Smart Shopping Cart

Palash Rathod, Preet Shah, Siddharth Sanghavi and Narendra Shekokar, Department of Computer Engineering, Dwarkadas J. Sanghvi College of Engineering, Mumbai, India

ABSTRACT

In today’s era, people look for ways to minimize effort in quotidian tasks like shopping. Shopping carts are meant to aid the customers to carry huge loads. However, with some modifications, we can revolutionize the way we use them. We propose a human following smart shopping cart to provide a hassle-free, costeffective shopping experience for the users in supermarkets and marts. Many attempts have been made to fulfil this task, but what these methods lacked was the freedom of movement of the cart to move in any direction according to the user. Our approach aims to grant this freedom to the cart by making sure that it follows the shopper even if he/she is not walking in a straight line. As an adjunct to this, a web user interface for keeping a track of the products in the cart will enhance the shopping experience for the users.

KEYWORDS

Smart Cart, Internet of Things, Magnetometer, RFID, MQTT.

Survey of Fake News Detection Techniques

Deshmukh Ashwinil and Dr.Sharvari Govilkar, PIillai's College of Engineering PCE, new Panvel, India

ABSTRACT

Fake News Detection is a long-standing problem faced by most of us through online or offline news sources. Large numbers of tools are present worldwide to detect fake news. However, for full success, detection of fake news requires lots of research on existing tools, an add-on to the browser, etc. Fake news could be in any form, half fake, satire, mostly fake in text or any multimedia format. Here the report presents a detailed study of fake news with its types, variations, and detailed discussion of each fake news category along with tools and existing methods used. In this paper, we will discuss some widely used fake news detection methods. Further, we analyze different features and drawbacks of Fake news detection methods and tools. There are many issues, and challenges still exist with many existing methods that are discussed in this report. Fake news detection is an evolving research problem that needs to be addressed with proper techniques and methods achieving efficiently and accurately.

KEYWORDS

Social Media, news data

Data Mic Calculation Method Based on Random Window

Zongchao Huang and Zhaogong Zhang, College of computer science and technology, Heilongjiang University, Harbin, China

ABSTRACT

Exploring the hidden correlation between data in unsupervised learning has always been a hot research topic. Reshef et al. providing the theory of MIC algorithm of maximum information coefficient for the first time in 2011, which further deepened the analysis of statistical measurement algorithm between variables However, the measurement standard is not suitable for detecting the correlation between two variables in a large data set. This paper proposes an algorithm for calculating the MIC value of data sets based on the idea of random window. By calculating the MIC value of data in a limited number of random windows that are not repeated, the W_MIC algorithm is proposed. Finally, the MIC value was compared with that of the original algorithm. The experimental results showed that the MIC method based on random window data sampling retained the maximum accuracy and greatly reduced the calculation time.

KEYWORDS

Correlation detection, MIC, random window

Intrusion Detection Based on Big Data Fuzzy Analytics

Hajer Bourass¹ and Farah Jemili², ¹Modeling of Automated Reasoning, University of Sousse, Hammam Sousse 4011, Tunisia and ²Universite de Sousse, ISITCom MARS Research Laboratory LR17ES05 Hammam Sousse 4011, Tunisia

ABSTRACT

In today’s world, Intrusion Detection System (IDS) is one of the significant tools used to the improvement of network security, which can detect the attacks or the abnormal access data. In the most of the existing IDS, the adoption of anomaly detection is a disadvantage such as high false alarm rate and low detection rate. For the IDS, there are several classifier algorithms such as Support Vector Machine (SVM) and Fuzzy C-Means (FCM). This paper proposes an Intrusion Detection System based on big data fuzzy analytics, Fuzzy C-Means (FCM) method is used to cluster and classify the pre-processed training dataset. The CTU-13 dataset and the UNSW-NB15 are used to proof the feasibility of the method. The performance evaluation is based on: accuracy, precision, detection rates, and false alarms.

KEYWORDS

Intrusion detection, IDS, Machine Learning, Apache Spark, Big Data, CTU-13, UNSW-NB15, Feature Selection, FCM clustering.