MACHINE LEARNING APPROACH FOR THE CLASSIFICATION OF VIOLENT ATTACK BASED ON FABRIC SENSORS

摘要

There are a number of women safety devices in the market today. Women who suffer these atrocities are even denied basic human rights, as set out in the Criminal Code. Women who are not as fit (physically) as men need to be protected from the evils of society. The introduction of actions and procedures for healthier women is not adequate and needs to be well improved. However, these devices are not foolproof. The summary of this paper is a partial result of a challenging problem faced during design and construction a fool-proof smart jacket for women's safety using fabric sensors. The smart jacket is envisioned for the women to wear in all occasions. An alert message and subjects1 geolocation are sent to pre-assigned phone number if the subject is faced with violent situation. The jacket consists of fabric sensors, accelerometer, gyroscope and magnetometer, which are strategically placed to record maximum variations in signal for minimum movement in subject's body. The primary challenge is not in design or construction of a jacket, but in accurately classifying violent activity from animated activity. Both violent activity and animated activities have commonality in sensor excitation [1]. There are subtle differences which need to be extracted to train a machine learning algorithm to learn these particular patterns. Multivariate regression analysis (MRA) is used to highlight explanatory variables that can produce better convergence. With MRA results signifying a high degree of noise in the data, a newer approach, wherein, the ordinality of the data was ignored and cardinality of the data was considered for analysis [2]. Machine learning approaches are then used to build a model using features extracted through MRA for the classification of different activities. This study presents that support vector machine shows the better classification accuracy, computation time as compare with other algorithms.