Transactions on Machine Learning and Artificial Intelligence

Fall detection system based on BiLSTM neural network

2019-11-09T06:23:45+00:00

The purpose of this article is to analyze the characteristics of human fall behavior to design a fall detection system. The existing fall detection algorithms have problems such as poor adaptability, single function and difficulty in processing large data and strong randomness. Therefore, a long-term and short-term memory recurrent neural network is used to improve the effect of falling behavior detection by exploring the internal correlation between sensor data. Firstly, the serialization representation method of sensor data, training data and detection input data is designed. The BiLSTM network has the characteristics of strong ability to sequence modeling and it is used to reduce the dimension of the data required by the fall detection model. then, the BiLSTM training algorithm for fall detection and the BiLSTM-based fall detection algorithm convert the fall detection into the classification problem of the input sequence; finally, the BiLSTM-based fall detection system was implemented on the TensorFlow platform. The detection and analysis of system were carried out using a bionic experiment data set which mimics a fall. The experimental results verify that the system can effectively improve the accuracy of fall detection to 90.47%. At the same time, it can effectively detect the behavior of Near-falling, and help to take corresponding protective measures.

Bioremediation of Hydrocarbon Contaminated Soil: Assessment of Compost Manure and Organic Soap

2019-11-09T06:23:41+00:00

This study was carried out to investigate the effect of compost manure and organic soap on hydrocarbon degradation in petroleum products contaminated soil. 10 kg of top soil collected at a depth of 0-20 cm, air dried and sieved, were poured into plastic containers. The soil samples were was pounded with 1 L of spent engine oil, 1 L of kerosene, 1 L of petrol and 1 L of diesel daily for five days. The containers were placed under natural environmental conditions for three weeks to enable full acclimatization of the petroleum products with the soil. A completely randomized design comprising T1 (Polluted soil without treatment ‘control’); T2 (10 kg contaminated soil + 500 g organic soap); T3 (10 kg contaminated soil + 500 g compost manure); and T4 (10 kg contaminated soil + 500 g compost manure + 500 g organic soap) was used for this study. Some physical characteristics (soil porosity and specific gravity) and Total Hydrocarbon Content (THC) of the soil samples were tested for, after the full acclimatization of the soil samples, and at the end of the 10 week experimental period, in accordance with standard methods. Results of the study showed that addition of the compost manure and organic soap the contaminated soil samples significantly (p ≤0.05) degraded the THC, and improved the soil physical characteristics. The result showed that the combination of compost manure and organic soap gave the best remediation result (from 957.21 mg/kg to 154.36 mg/kg), followed by organic soap (from 957.21mg/kg to 203.61 mg/kg), and then compost manure (from 957.21 mg/kg to 262.03 mg/kg). At the end of the experimental period, vegetative growth was observed in the treated soil samples; whereas, in the control soil samples vegetative growth was absent. Results obtained from this study have shown that amending petroleum products contaminated soils with compost manure and organic soap will enhance remediation of petroleum products contaminated sites.

A survey of Emerging Techniques in Detecting SMS Spam

2019-11-09T06:23:43+00:00

In the past years, spammers have focused their attention on sending spam through short messages services (SMS) to mobile users. They have had some success because of the lack of appropriate tools to deal with this issue. This paper is dedicated to review and study the relative strengths of various emerging technologies to detect spam messages sent to mobile devices. Machine Learning methods and topic modelling techniques have been remarkably effective in classifying spam SMS. Detecting SMS spam suffers from a lack of the availability of SMS dataset and a few numbers of features in SMS. Various features extracted and dataset used by the researchers with some related issues also discussed. The most important measurements used by the researchers to evaluate the performance of these techniques were based on their recall, precision, accuracies and CAP Curve. In this review, the performance achieved by machine learning algorithms was compared, and we found that Naive Bayes and SVM produce effective performance.

Artificial Soul Optimization - An Invention

2019-11-09T06:23:38+00:00

The Soul is eternal and exists even after death of a person or animal. The main idea that is captured in this work is that soul continues to exist and takes a different a body after the death. The primary goal of this work is to invent a new field titled "Artificial Soul Optimization (ASO)". The term "Artificial Soul Optimization" is coined in this paper. All the Optimization algorithms which are proposed based on Artificial Souls will come under "Artificial Soul Optimization" Field (ASO Field). In the Particle Swarm Optimization and Artificial Human Optimization, the basic entities in search space are Artificial Birds and Artificial Humans respectively. Similarly, in Artificial Soul Optimization, the basic entities in search space are Artificial Souls. In this work, the ASO Field concepts are added to Particle Swarm Optimization (PSO) algorithm to create a new hybrid algorithm titled "Soul Particle Swarm Optimization (SoPSO). The proposed SoPSO algorithm is applied on various benchmark functions. Results obtained are compared with PSO algorithm. The World's first Hybrid PSO algorithm based on Artificial Souls is created in this work.