Robust Phishing Detection Against Adversaries

Phishing websites have grown more recently than ever, and they become more intelligent, even against well-designed phishing detection techniques. Formerly, we have proposed in the literature a state-of-the-art URL-exclusive phishing detection solution based on Convolutional Neural Network (CNN) model, which we referred as PUCNN model. Phishing detection is adversarial as the phisher may attempt to avoid the detection. This adversarial nature makes standard evaluations less useful in predicting model performance in such adversarial situations. We aim to improve PUCNN by addressing the adversarial nature of phishing detection with a restricted adversarial scenario, as PUCNN has shown that an unrestricted attacker dominates. To evaluate this adversarial scenario, we present a parameterized text-based mutation strategy used for generating adversarial samples. These parameters tune the attacker’s restrictions. We have focused on text-based mutation due to our focus on URL-exclusive models. The PUCNN model generally showed robustness and performed well when the parameters were low, which indicates a more restricted attacker.

Stock Prediction Using Machine Learning

The Trend of stock price prediction is becoming more popular than ever. Share market is difficult to predict due to its volatile nature. There are no rules to follow to predict what will happen with the stock in the future. To predict accurately is a huge challenge since the market trend always keep changing depending on many factors. The objective is to apply machine learning techniques to predict stocks and maximize the profit. In this work, we have shown that with the help of artificial intelligence and machine learning, the process of prediction can be improved. While doing the literature review, we realized that the most effective machine learning tool for this research include: Artificial Neural Network (ANN), Support Vector Machine (SVM), and Genetic Algorithms (GA). All categories have common and unique findings and limitations. We collected data for about 10 years and used Long Short-Term Memory (LSTM) Neural Network-based machine learning models to analyze and predict the stock price. The Recurrent Neural Network (RNN) is useful to preserve the time-series features for improving profits. The financial data High and Close are used as input for the model.

Timetable and Headway Adherence Assessment using the Gini Coefficient and the Lorenz Curve

Adherence to the schedule is of prime importance in public transport. This paper presents a specific application of the Gini coefficient, well known indicator in economics, for the headway adherence assessment. The paper shows that Lorenz curve, which is usually used to define mathematically the Gini coefficient, is a good indicator of the users' waiting time when it is based on the bus schedule. When it is computed on the basis of the ratio of observed headway to the schedule, it is a powerful visual tool that can be used by operators to detect the existence of irregularities on a bus line at a glance. An equation gives, in an idealistic case, the impact of any single traffic disturbance on the Gini coefficient, making this coefficient comprehensive. A detailed analysis is developed, based on the bus proportions according to the headway adherence level. These proportions are obtained from new indices coming from the derivative of the Lorenz curve. The values of these indexes alert the operator of any adherence disturbance. The examination of the Lorenz curve takes more time, but is worthwhile, giving the types of the irregularities The application of these indicators is carried on real-time data from the New Delhi bus network.

Software Identifier Naming Conventions & Dictionary

Software developers have been presented with so many tools meant to assist then during the development process. Tools like autocomplete, intelli-sense, linters, and other static analysis solutions. All such tools have one underlying goal, to promote productivity and improve quality. Much research has been conducted on the topic of software quality and its direct benefits both during and after the development cycle. Various methods of measuring and improving quality in software products have been implemented at a grand scale. However, software developers are still left with the choice of implementation details. One such detail is the choice of identifier names in the code written. Few publications have focused on conventions, guides, or best-practices on the topic of identifiers naming choices (not to be confused with coding styles). Much time and energy is misused by developers while choosing an appropriate identifier name, as well as by other developers later on when trying to understand the choice made by their colleagues. By aggregating and compiling a list of readily available identifier names that developers can choose from, will allow them to focus on other keys aspects of development

A High Performance Global Routing Algorithm on Julia Parallel Computing Platform

o keep pace with the design requirements of Integrated Circuits (ICs), parallel processing is adopted. The path to be routed between two nodes may or may not be dependent on the previously routed paths. The solution requires careful attention in distributing the nets to be routed to different processors. Previous work on allocating the tasks to processors has been quite successful, reporting upto 3x improvement on 4 cores and 5x improvement on 8 core machine. The advantage of increasing the number of cores diminishes with each added processor and the challenge lies in being able to maintain the improvement per added core. The existing techniques of distributing the nets cannot provide additional advantage of using more than 8 cores. This paper improves the work on parallelizing global routing using a technique of balancing the load on the processors for better utilization of the resources. A relatively new budding platform Julia has been used which provides the ease of programming while maintaining the performance of the C language. Technique used in this paper has enabled to use 16 cores with routing solutions obtained in 0.8 minutes achieving 12.5 times speedup compared to sequential processing on a single core

Continuous Software Engineering and Unit Testing: From Theory to Practice

With the Agile development approach, the software industry has moved to a more flexible and continuous Software Development Life Cycle (SDLC), which integrates the stages of development, delivery and deployment. This trend has exposed a tendency of increasing reliance on both unit testing and test automation for the fundamental quality-activities during the code development. To implement Continuous Software Engineering (CSE), it is vital to assure that unit-testing activities are an integral and well-defined part of a continuous process. This paper focuses on the initial role of actual testing – viewing unit testing as a quality indicator during the development life cycle. We review the definition of unit-testing from the CSE world, and describe a qualitative study in which we examined implementation of unit testing in three software companies that recently migrated to CSE methodology. The results from the qualitative study corroborate our argument that under the continues approach, quality-based development practices such as unit testing are of increasing importance, lacking common set of measurements and KPI's. A possible explanation to this may be the role of continuous practices as well as unit testing in the software engineering curriculum

A Tiling Algorithm-Based String Similarity Measure

This paper describes a similarity measure for strings based on a tiling algorithm. The algorithm is applied to a pair of proteins that are described by their respective amino acid sequences. The paper also describes how the algorithm can be used to find highly conserved amino acid sequences and examples of horizontal gene transfer between different species

Evaluating Inclusive Campus Environment Design Criteria Using CFPR and FANP Methodologies

University campuses bring together individuals from different socio-cultural backgrounds. At the same time, university campuses contribute to the personal and intellectual development of individuals and serve as a socialization area. Campuses create vitality with their social, cultural, economic, and spatial effects. In this paper, we study for evaluating inclusive campus environment design criteria using the Fuzzy Analytical Network Process (FANP) and Consistent Fuzzy Preference Relations (CFPR) techniques, which are two Multi- Criteria Decision Making (MCDM) methods. Seven Inclusive Campus Environment Design Criteria are “Land Use Organization”, “Compactness”, “Connectivity”, “Configuration”, “Living campus”, “Greens” and “Context”. The major contribution of our study is to prioritize inclusive campus environment design criteria by using numerical methods from the decision maker's perspective. According to the authors’ knowledge, this will be the first interdisciplinary study to use MCDM methods for evaluating inclusive campus environment design criteria. Additionally, the results of both methodologies are compared

Forecasting of Visitors Arrived in Taiwan for Tourism Supply Chain Demand using Big Data

The fast development of Information and Communication Technology, generate, collect and operate a large amount of data, which is termed big data. The search queries in web search engines can be retrieved by visitors to obtain useful infor-mation for the selected next visiting destinations. Google Trends on Google search engine can evaluate and compare how many times users are searching for specific terms or topics. Otherwise, economic factors, covering income, the rela-tive prices, and relative exchange rate usually influence the international tourist demand. However, there are different conclusions in different settings. Accord-ingly, this work presents the ARIMAX model for modelling and forecasting numbers of international tourists visiting Taiwan from Japan for different pur-poses and provides an analysis of the effects of big data and economic factors. The results can contribute to the decision makers of the tourism industry in Taiwan

Road Accidents Analysis Using Comparative Study & Application of Machine Learning Algorithms

Model predictive contrTraffic Collisions are one of the major sources of deaths, injuries & property damage every year. Road accidents are one of the most difficult real world problems to tackle with, due to its high order of unpredictability. The persistence as well as existence of this problem may be prevalent to a different degree for each & every place. The consequences of this may result in loss of human life & capital. To avoid this, every place needs to tackle the problem with a customized approach depending on the causes that are responsible for the accidents. Even in today's world, where the mass operation of autonomous vehicles is still grim or out of sight, the possibility of predicting a road accident before it takes place, is practically impossible. The only idea or approach that can help to decrease the number of road accidents, is to analyze the reasons that lead to these accidents. The concepts of Data Analysis, Data Visualization & Machine Learning help to tackle real world problems, by exploring & deriving valuable insights, which in turn help in taking measures to solve the targeted problem & drive business growth. In this research study, the dataset pertaining to road mishaps that occurred in UK over time period 2005 - 2015 will be analyzed using these concepts. The defined approach can help the concerned authorities & respective government, to take every possible step & amendment, & hence mitigate the identified causes & scenarios that lead to road accidents.