Hybrid Machine Learning: A Tool to Detect Phishing Attacks in Communication Networks

Phishing is a cyber-attack that uses disguised email as a weapon and has been on the rise in recent times. Innocent Internet user if peradventure clicking on a fraudulent link may cause him to fall victim of divulging his personal information such as credit card pin, login credentials, banking information and other sensitive information. There are many ways in which the attackers can trick victims to reveal their personal information. In this article, we select important phishing URLs features that can be used by attacker to trick Internet users into taking the attacker’s desired action. We use two machine learning techniques to accurately classify our data sets. We compare the performance of other related techniques with our scheme. The results of the experiments show that the approach is highly effective in detecting phishing URLs and attained an accuracy of 97.8% with 1.06% false positive rate, 0.5% false negative rate, and an error rate of 0.3%. The proposed scheme performs better compared to other selected related work. This shows that our approach can be used for real-time application in detecting phishing URLs.

A Thermal-Based Fuel-Prediction Method for Intelligent Fire Extinguisher in an Indoor Environment

Classification of fuel in the early stage of fire is important to choose the appropriate type of extinguisher for extinguishing fire. This paper proposes a method of fuel prediction based on heat information for intelligent fire extinguisher in an indoor environment. Fire flame in the early stage is first detected based on patterns of differences between consecutive thermal image frames in which temperature grows up rapidly and reveals a sharp positive slope. Then candidate flame boundaries are detected in the thermal image frames during the early stage, and boundary matching is performed among the frames. These matched boundaries are classified as fire flame and fuel class based on LSTM (Long short-term memory) for extinguisher selection. Experiments were performed with 300 samples for classification into four classes of fuel, and the results based on 9:1 training and testing ratio showed 92.142% accuracy.

A Study on Usability and Motion Sickness of Locomotion Techniques for Virtual Reality

Virtual Reality (VR) is widely used today in both research and entertainment. The continuous growth of this technology makes VR consumer hardware now available for masses. The new trend in the next generation of VR devices is a VR headset and controllers with inside-out technology. These VR devices will become an important basis for the future of VR applications. Virtual travel or locomotion inside VR experiences is the important part in the VR application development, which affected to the users preference. The goal of this research is to study the difference of locomotion in VR with new trend devices consisting of VR headset and controllers without using other accessories. Three locomotion techniques: controller-based, motion-based and teleportation-based were used to analyze the differences. The VR scene with virtual environments was created to use in the experiment where the users have to move with different locomotion technique. The Usability Questionnaire (UQ) is used to evaluate the usability value of each locomotion technique, while Simulator Sickness Questionnaire (SSQ) is used to assess the motion sickness value. The results showed that the usability (p-value=0.02007) and motion sickness (p-value=0.00014) of all locomotion techniques are different and the usability affected to the user preference. The conclusions of VR locomotion studies were discussed with the limitations of the study and the future work for this research.

Damaged Vehicle Parts Detection Platforms using Deep Learning Technique

Automatic vehicle damage detection platform can increase the market value of car insurance. The es- timation process is usually manual and requires hu- man experts and their time to evaluate the damage cost. Intelligent Vehicle Accident Analysis (IVAA) system provides an artificial intelligence as a service (AIaaS) for building a system that can automatically assess vehicle parts’ damage and severity level. The insurance company can adopt our service to build the application to speedup the claiming process. There are four main elements in the service system which support four stakeholders in an insurance company: insurance experts, data scientists, operators and field employees. Insurance experts utilize the data label- ing tool to label damaged parts of a vehicle in a given image as a training data building process. Data scientists iterate to the deep learning model build- ing process for continuous model updates. Opera- tors monitor the visualization system for daily statis- tics related to the number of accidents based on lo- cations. Field employees use LINE Official integra- tion to take a photo of damaged vehicle at the acci- dent site and retrieve the repair estimation. IVAA is built on the docker image which can scale-in or scale- out the system depend on utilization efficiently. We deploy the Faster Region-based convolutional neural network, along with residual Inception network to lo- calize the damage region and classify into 5 damage levels for a vehicle part. The accuracy of the localiza- tion is 93.28 % and the accuracy of the classification is 98.47%.

Identification of Soft Fall based on Falling State Occurrences

Fall detection techniques for helping the elderly were developed based on identifying falling states using simulated falls. However, some real-life falling states were left undetected, which led to this work on analysing falling states. The aim was to find the differences between active daily living and soft falls where falling states were undetected. This is the first consideration to be based on the threshold-based algorithms using the acceleration data stored in an activity database. This study addresses soft falls in addition to the general falls based on two falling states. Despite the number of false alarms being higher rising from 18.5% to 56.5%, the sensitivity was increased from 52% to 92.5% for general falls, and from 56% to 86% for soft falls. Our experimental results show the importance of state occurrence for soft fall detection, and will be used to build a learning model for soft fall detection.

Cascading Models of CNN and GRU with Autoencoder Loss for Precipitation Forecast in Thailand

It is a crucial task to accurately forecast precipitation, especially rainfall in Thailand, since it relates to flood prevention and agricultural planning. In our prior work, we have presented a model based on deep learning approach; however, its performance is still limited due to two main issues. First, there is an imbalance issue, where most rainfall is zero or no rain because Thailand has short rainy season. Second, predicted rainfall is still underestimated since moderate and heavy rainfall cases barely occurs. In this paper, we propose an enhanced deep learning model to forecast rainfall in Thailand. Our model is a cascading of CNN and GRU along with exogenous variables, i.e., temperature, pressure, and humidity. There are two stages in our model. First, CNN is specialized for classifying rain and non-rain events. In this stage, an imbalanced issue is alleviated by applying “focal loss”. Second, GRU is responsible for forecasting rainfall. Its predicted range is lifted using “autoencoder loss”. The experiment was conducted on hourly rainfall dataset between 2012 and 2018 obtained from a public government sector in Thailand. The results show that our enhanced model outperforms ARIMA and CNN-GRU in terms on RMSE of most regions in Thailand.

Neuro-fuzzy Architecture for Handover Decision Making in Emerging Heterogeneous Networks

In emerging heterogeneous networks, seamless vertical handover is a critical issue. There must be a trade-off between the handover decision delay and accuracy. This paper’s concern is to contribute to reliable vertical handover decision making that makes a trade-off between complexity and effectiveness. So, the paper proposes a neuro-fuzzy architecture that joints the capacity of learning of the artificial neural networks with the power of linguistic interpretation of the fuzzy logic. The architecture can learn from experience how executing a handover to a particular access network affects the quality of service. Simulation results reveal that this architecture is fast, enhances the overall performance and reliability better than the fuzzy logic-based approach.

Evaluation of a Textile PIFA for Wearable IoT Application and its Challenges

The ever increasing use of body-worn systems in the Internet of Things application such as needs better antenna subsystem designs compatible with its requirements. Several challenges limiting the performance of a body-worn system, from materials, and environmental conditions to the effects of on body application and its hazards are discussed. As a test case, a flexible textile planar inverted-F antenna is presented and discussed. The choice of this topology is due to its simplicity in design and fabrication, relatively broad bandwidth and the presence of a rear ground plane, which minimizes the impacts of the human body on the antenna performance. It is designed on a felt substrate, whereas Aaronia-shield conductive textile is utilized as its conductive parts (radiator, shorting wall and ground plane). The antenna performance are studied in two cases, first in free space and then in bent conditions in the close proximity to the human body. The influence of the relative humidity on the textile antenna performance is also investigated numerically. Simulated and measured results indicated good agreements. Finally, the proposed antenna is integrated with a transceiver module and evaluated on the body in practice. Its wireless link quality is assessed in an indoor laboratory.

Simulation Inquisition of Noise Dissolving Algorithm Hinge on TTSD filter for RIIN Situation

This primary aim of this philosopher paper investigates the efficacy of the noise dissolving algorithm hinge on TTSD (Triple Threshold Statistical Detection) filter that has been originated since 2018 is one of the highest efficacy for dissolving RIIN (Random-Intensity Impulse Noise), exclusively at dense distribution. As a results, there are three essential contributions: the exhaustive explanation of the TTSD filter algorithm and its computation examples, the calculation simulation of noise apprehension correctness and overall comparative simulation of noise dissolving effectiveness. For TTSD filter, three malleable offsets that are the complementary requirement are employed in the TTSD filter that can adequately resolve the limitation of the antecedent noise dissolving algorithms. The first malleable offset is calculated for determining the noise characteristic of all elements by using the mathematical verification. Next, the second malleable offset is calculated for determining the another noise characteristic by using the normal distribution mathematical verification (the average value and standard deviation value). Later, the third malleable offset is calculated for determining the another noise characteristic by using the quartile mathematical verification (median value). In the simulation inquisition, the bountiful standard portraits that are desecrated by RIIN (Random Intensity Impulse Noise) with many dense distributions are experimented by noise dissolving algorithm hinge on TTSD in both noise segregation and noise dissolving perspective.

Integrated Healthcare Services for Ageing in Place

This study investigates how healthcare practitioners handle significant circumstancesof providing medical assistance and treatments to patients and what challenges theyface. Drawing on key healthcare stakeholders and mixed smart living methods, wedevelop a guideline service protocol for Internet of Things (IoT) solution to helphealthcare stakeholders in coping with operational difficulties. IoT technology is one ofthe key determinants that empowers healthcare professionals to achieve their tasks,and our goal is to study the functions that provides to local citizens, especially olderpeople, and to evaluate how the functions and platform could assist corporatecompliance policies to increase the efficiency of healthcare service. Our fieldexperiments have indicated a need to educate healthcare users about IoT applicationthat provide advantages in decision making. In addition, our research has explored andevaluated the impacts and factors that influence the development and collaboration byallowing workflows of healthcare stakeholders and by following integrated smart livingplatform and required service protocol.