Pub Date : 2020-07-01DOI: 10.1109/ComPE49325.2020.9200139
Pritom Gogoi, Manpa Barman, Mahendra Deka, Upasana Rajkonwar, Rhittwikraj Moudgollya
The involvement of embedded systems and computer vision is increasing day by day in various segments of consumer market like industrial automation, traffic monitoring, medical imaging, modern appliance market, augmented reality systems, etc. These technologies are bound to make new developments in the domain of commercial and home security surveillance. Our project aims to make contributions to the domain of video surveillance by making use of embedded computer vision systems. Our implementation, built around the Raspberry Pi 4 SBC aims to utilize computer vision techniques like motion detection, face recognition, object detection, etc to segment the region of interest from the captured video footage. This technique is superior as compared to traditional surveillance systems as it requires minimum human interaction and intervention at the control room of such security systems. The proposed system is capable of sensing suspicious events like detection of an unknown face in the captured video or motion detection/object detection in a closed section of a building. Moreover, with the help of the turret mechanism built using servo motors, the camera integrated in the system is capable of having 360◦ rotation and can track a detected face or object of interest within its range. Apart from automated tracking, the system can also be manually controlled by the operator.
{"title":"Object Detection and Tracking Turret based on Cascade Classifiers and Single Shot Detectors","authors":"Pritom Gogoi, Manpa Barman, Mahendra Deka, Upasana Rajkonwar, Rhittwikraj Moudgollya","doi":"10.1109/ComPE49325.2020.9200139","DOIUrl":"https://doi.org/10.1109/ComPE49325.2020.9200139","url":null,"abstract":"The involvement of embedded systems and computer vision is increasing day by day in various segments of consumer market like industrial automation, traffic monitoring, medical imaging, modern appliance market, augmented reality systems, etc. These technologies are bound to make new developments in the domain of commercial and home security surveillance. Our project aims to make contributions to the domain of video surveillance by making use of embedded computer vision systems. Our implementation, built around the Raspberry Pi 4 SBC aims to utilize computer vision techniques like motion detection, face recognition, object detection, etc to segment the region of interest from the captured video footage. This technique is superior as compared to traditional surveillance systems as it requires minimum human interaction and intervention at the control room of such security systems. The proposed system is capable of sensing suspicious events like detection of an unknown face in the captured video or motion detection/object detection in a closed section of a building. Moreover, with the help of the turret mechanism built using servo motors, the camera integrated in the system is capable of having 360◦ rotation and can track a detected face or object of interest within its range. Apart from automated tracking, the system can also be manually controlled by the operator.","PeriodicalId":6804,"journal":{"name":"2020 International Conference on Computational Performance Evaluation (ComPE)","volume":"98 1","pages":"792-796"},"PeriodicalIF":0.0,"publicationDate":"2020-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79223635","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-07-01DOI: 10.1109/ComPE49325.2020.9200143
Lakshmi Dutta, Dushmanta Kumar Das
This paper presents a linear model predictive control scheme for a magnetic levitation system (Maglev). The Maglev dynamics involve nonlinearity, highly unstable, makes it more challenging to design a suitable control algorithm. The objective of this work is to control the position of a ferromagnetic ball in the air space of the nonlinear system. In this research, the proposed controller is designed based on the linear prediction model, which is obtained by linearizing the plant around a known operating point. The effectiveness of the proposed linear model predictive control algorithm is verified in the simulation environment. For comparative analysis, the performance of the proposed controller is compared with a 1-DOF PID control technique [1] and found better results.
{"title":"A Linear Model Predictive Control design for Magnetic Levitation System","authors":"Lakshmi Dutta, Dushmanta Kumar Das","doi":"10.1109/ComPE49325.2020.9200143","DOIUrl":"https://doi.org/10.1109/ComPE49325.2020.9200143","url":null,"abstract":"This paper presents a linear model predictive control scheme for a magnetic levitation system (Maglev). The Maglev dynamics involve nonlinearity, highly unstable, makes it more challenging to design a suitable control algorithm. The objective of this work is to control the position of a ferromagnetic ball in the air space of the nonlinear system. In this research, the proposed controller is designed based on the linear prediction model, which is obtained by linearizing the plant around a known operating point. The effectiveness of the proposed linear model predictive control algorithm is verified in the simulation environment. For comparative analysis, the performance of the proposed controller is compared with a 1-DOF PID control technique [1] and found better results.","PeriodicalId":6804,"journal":{"name":"2020 International Conference on Computational Performance Evaluation (ComPE)","volume":"97 1","pages":"039-043"},"PeriodicalIF":0.0,"publicationDate":"2020-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79228012","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-07-01DOI: 10.1109/ComPE49325.2020.9200082
Tanaya Das, Dhruba Jyoti Sut, Vishal Gupta, Lakhyajit Gohain, Priyanka Kakoty, N. M. Kakoty
This paper reports development of a robot that can sense the presence of hazardous gases in the environment. The robot aims at detection of hazardous gases and mapping global positioning system (GPS) locations of the detected gases to the navigation terrain in real-time. These information was transmitted to a hand-held device in a remote location for exploration of gas types, which holds promise for disaster management. The robot was equipped with a module of gas sensors, human detection sensor, GPS module and obstacle detection sensors in a coherent system. While navigating with collision avoidance to obstacles, the robot can transmit information about the presence of hazardous gases and human being in the area of navigation. It was tested in an uneven terrain to recognize the presence of hazardous gases like carbon dioxide, liquefied petroleum gas, vaporized alcohol gas vis-a-vis ambient gases in real-time. A ´ neural network-based classifier was implemented to recognize the gases with an average accuracy of 98%.
{"title":"A Mobile Robot for Hazardous Gas Sensing","authors":"Tanaya Das, Dhruba Jyoti Sut, Vishal Gupta, Lakhyajit Gohain, Priyanka Kakoty, N. M. Kakoty","doi":"10.1109/ComPE49325.2020.9200082","DOIUrl":"https://doi.org/10.1109/ComPE49325.2020.9200082","url":null,"abstract":"This paper reports development of a robot that can sense the presence of hazardous gases in the environment. The robot aims at detection of hazardous gases and mapping global positioning system (GPS) locations of the detected gases to the navigation terrain in real-time. These information was transmitted to a hand-held device in a remote location for exploration of gas types, which holds promise for disaster management. The robot was equipped with a module of gas sensors, human detection sensor, GPS module and obstacle detection sensors in a coherent system. While navigating with collision avoidance to obstacles, the robot can transmit information about the presence of hazardous gases and human being in the area of navigation. It was tested in an uneven terrain to recognize the presence of hazardous gases like carbon dioxide, liquefied petroleum gas, vaporized alcohol gas vis-a-vis ambient gases in real-time. A ´ neural network-based classifier was implemented to recognize the gases with an average accuracy of 98%.","PeriodicalId":6804,"journal":{"name":"2020 International Conference on Computational Performance Evaluation (ComPE)","volume":"32 1","pages":"062-066"},"PeriodicalIF":0.0,"publicationDate":"2020-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"84708299","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-07-01DOI: 10.1109/ComPE49325.2020.9200008
Hara Gobinda Naskar, A. Maity
A biquad Gm-C programmable band-pass filter has been designed for wireless applications. The filter is designed by cascading four biquad stages with each stage consisting of a parallel combination of an active resistor, capacitor and active inductor. Nauta trans-conductor stage is used as a trans-conductance block which is also used to realize the active resistor. Karsilayan active inductors are used in place of gyrator-C active inductors to reduce the power consumption. A new tuning technique is devised to digitally control the centre frequency of the filter ranging from 910 MHz to 2.09 GHz with a pass-band gain of ~40dB. In the proposed tuning scheme, the Karsilayan active inductor has been varied instead of the capacitor. The power consumption of the filter is relatively lower (close to 56 mW) as compared to other high frequency filters operating in the same frequency range. The design has been carried out in 180-nm standard CMOS technology.
{"title":"A Gm-C Biquad Programmable Band Pass Filter for Wireless Applications","authors":"Hara Gobinda Naskar, A. Maity","doi":"10.1109/ComPE49325.2020.9200008","DOIUrl":"https://doi.org/10.1109/ComPE49325.2020.9200008","url":null,"abstract":"A biquad Gm-C programmable band-pass filter has been designed for wireless applications. The filter is designed by cascading four biquad stages with each stage consisting of a parallel combination of an active resistor, capacitor and active inductor. Nauta trans-conductor stage is used as a trans-conductance block which is also used to realize the active resistor. Karsilayan active inductors are used in place of gyrator-C active inductors to reduce the power consumption. A new tuning technique is devised to digitally control the centre frequency of the filter ranging from 910 MHz to 2.09 GHz with a pass-band gain of ~40dB. In the proposed tuning scheme, the Karsilayan active inductor has been varied instead of the capacitor. The power consumption of the filter is relatively lower (close to 56 mW) as compared to other high frequency filters operating in the same frequency range. The design has been carried out in 180-nm standard CMOS technology.","PeriodicalId":6804,"journal":{"name":"2020 International Conference on Computational Performance Evaluation (ComPE)","volume":"74 1","pages":"396-401"},"PeriodicalIF":0.0,"publicationDate":"2020-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79670002","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-07-01DOI: 10.1109/ComPE49325.2020.9200014
Utpal Barman, Ridip Dev Choudhury, Asif Ekbal Hussain, Mridul Jyoti Dahal, Puja Barman, M. Hazarika
Weather Forecasting is a serious issue in agriculture, especially in Assam and Meghalaya. The productivity of agriculture is dependent on rainfall. This paper forwards a comparative assessment of Auto-Regressive, Moving Average, and Auto-Regressive Moving Average Model for rainfall in Assam and Meghalaya. A total of 117 years of rainfall data of Assam and Meghalaya division is collected from data.gov.in [11]. The models are implemented by visualizing the time series components of rainfall. The necessary investigations such as ACF, PACF, rolling mean, and ducky fuller tests are reported in the paper for the analysis of stationarity of time series. The evaluating parameters such regression score (0.73), mean absolute error (75.70), median absolute error (61.43), mean squared error (9396.09), mand root mean square error (96.93) select the ARMA model as the best model for the time series forecasting of Assam and Meghalaya Division.
{"title":"Comparative Assessment of AR, MA and ARMA for the Time Series Forecasting of Assam and Meghalaya Rainfall Division","authors":"Utpal Barman, Ridip Dev Choudhury, Asif Ekbal Hussain, Mridul Jyoti Dahal, Puja Barman, M. Hazarika","doi":"10.1109/ComPE49325.2020.9200014","DOIUrl":"https://doi.org/10.1109/ComPE49325.2020.9200014","url":null,"abstract":"Weather Forecasting is a serious issue in agriculture, especially in Assam and Meghalaya. The productivity of agriculture is dependent on rainfall. This paper forwards a comparative assessment of Auto-Regressive, Moving Average, and Auto-Regressive Moving Average Model for rainfall in Assam and Meghalaya. A total of 117 years of rainfall data of Assam and Meghalaya division is collected from data.gov.in [11]. The models are implemented by visualizing the time series components of rainfall. The necessary investigations such as ACF, PACF, rolling mean, and ducky fuller tests are reported in the paper for the analysis of stationarity of time series. The evaluating parameters such regression score (0.73), mean absolute error (75.70), median absolute error (61.43), mean squared error (9396.09), mand root mean square error (96.93) select the ARMA model as the best model for the time series forecasting of Assam and Meghalaya Division.","PeriodicalId":6804,"journal":{"name":"2020 International Conference on Computational Performance Evaluation (ComPE)","volume":"143 1","pages":"507-511"},"PeriodicalIF":0.0,"publicationDate":"2020-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"77311243","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-07-01DOI: 10.1109/ComPE49325.2020.9200172
Navarun Gupta, S. Patel, Lawrence V. Hmurick
The Wheatstone bridge is a very common and very important electric circuit, especially in the measurement of resistance/impedance corresponding to transducers and materials deformity. See Figure 1 . DC bridges are most often used for transducer measurements, both mechanical, material, and optical (1) , (2) , and (3) . AC bridges are often used to measure inductance or capacitance (4) , (5) . Both types of bridges require the voltage output of the bridge to be zero, and this is accomplished when the 4 arms of the bridge are balanced so as to pass the same current through the 2 left impedances as through the 2 right impedances, or in other words, the output voltage and current are zero.
{"title":"Using an Un-Balanced AC Wheatstone Bridge to Measure Capacitance and Inductance","authors":"Navarun Gupta, S. Patel, Lawrence V. Hmurick","doi":"10.1109/ComPE49325.2020.9200172","DOIUrl":"https://doi.org/10.1109/ComPE49325.2020.9200172","url":null,"abstract":"The Wheatstone bridge is a very common and very important electric circuit, especially in the measurement of resistance/impedance corresponding to transducers and materials deformity. See Figure 1 . DC bridges are most often used for transducer measurements, both mechanical, material, and optical (1) , (2) , and (3) . AC bridges are often used to measure inductance or capacitance (4) , (5) . Both types of bridges require the voltage output of the bridge to be zero, and this is accomplished when the 4 arms of the bridge are balanced so as to pass the same current through the 2 left impedances as through the 2 right impedances, or in other words, the output voltage and current are zero.","PeriodicalId":6804,"journal":{"name":"2020 International Conference on Computational Performance Evaluation (ComPE)","volume":"55 1","pages":"266-268"},"PeriodicalIF":0.0,"publicationDate":"2020-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90791032","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-07-01DOI: 10.1109/ComPE49325.2020.9200026
Uttara Biswas, J. K. Rakshit, M. Singh
In the medical diagnosis, urinanalysis has a significant role in the recognition of many infections, diabetes and kidney diseases. In this article, photonic crystal (PhC) micro ring resonator (MRR) is proposed for different hematological or renal disorders where the concentrations of different components of urine can be detected using refractive index (RI) variation of urine. The proposed sensor works on the principle of resonance wavelength shift due to analytes having different RI and is simulated using FDTD simulation. High sensitivity is achieved with the reported structure as 720 nm/RIU for various glucose concentrations, 830 nm/RIU for various albumin concentrations and 701 nm/RIU for different urea concentration in the urine and the structure would be beneficial for various sensing purpose.
{"title":"Photonic Crystal based Micro Ring Resonator Sensor Design for Urinanalysis","authors":"Uttara Biswas, J. K. Rakshit, M. Singh","doi":"10.1109/ComPE49325.2020.9200026","DOIUrl":"https://doi.org/10.1109/ComPE49325.2020.9200026","url":null,"abstract":"In the medical diagnosis, urinanalysis has a significant role in the recognition of many infections, diabetes and kidney diseases. In this article, photonic crystal (PhC) micro ring resonator (MRR) is proposed for different hematological or renal disorders where the concentrations of different components of urine can be detected using refractive index (RI) variation of urine. The proposed sensor works on the principle of resonance wavelength shift due to analytes having different RI and is simulated using FDTD simulation. High sensitivity is achieved with the reported structure as 720 nm/RIU for various glucose concentrations, 830 nm/RIU for various albumin concentrations and 701 nm/RIU for different urea concentration in the urine and the structure would be beneficial for various sensing purpose.","PeriodicalId":6804,"journal":{"name":"2020 International Conference on Computational Performance Evaluation (ComPE)","volume":"27 1","pages":"153-157"},"PeriodicalIF":0.0,"publicationDate":"2020-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"77882331","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-07-01DOI: 10.1109/ComPE49325.2020.9200071
Banita
Parkinson disease (PD) is a neurodegenerative disorder. Many common symptoms which may or may not indicate that patient is suffering from Parkinson Disease. In this study a novel rating scale has been introduced which helps to examine the level of Parkinson Disease but is not mandatory that a person having similar symptoms may surely suffering from Parkinson Disease. PD is an unsolved problem till date hence the study focuses on relevant features, drugs and common techniques used to detect or analyze PD. To overcome such problem different techniques will be used to study and analyze the early detection of PD. It can be analyzed with the help of deep understanding of Parkinson Disease. However presence of some common symptoms has not yet been described up to the mark to analyze the level of Parkinson Disease. Hence it is very challenging to detect early stage of Parkinson Disease. In study, work focuses on only confirmed symptoms of Parkinson Disease which doesn’t deals to any other disease completely. Study also focus on the Unified Parkinson’s Disease Rating scale (UPDRS) for Parkinson Disease along with respective symptoms. It includes the analysis in terms of medical science and computer applications for analyzing PD. Medication for PD has also been discussed in the study including wide literature survey which provide the clearance of the goal for treating PD. Proposed rating scale in the study is time efficient as compared with UPDRS.
{"title":"Detection of Parkinson’s Disease Using Rating Scale","authors":"Banita","doi":"10.1109/ComPE49325.2020.9200071","DOIUrl":"https://doi.org/10.1109/ComPE49325.2020.9200071","url":null,"abstract":"Parkinson disease (PD) is a neurodegenerative disorder. Many common symptoms which may or may not indicate that patient is suffering from Parkinson Disease. In this study a novel rating scale has been introduced which helps to examine the level of Parkinson Disease but is not mandatory that a person having similar symptoms may surely suffering from Parkinson Disease. PD is an unsolved problem till date hence the study focuses on relevant features, drugs and common techniques used to detect or analyze PD. To overcome such problem different techniques will be used to study and analyze the early detection of PD. It can be analyzed with the help of deep understanding of Parkinson Disease. However presence of some common symptoms has not yet been described up to the mark to analyze the level of Parkinson Disease. Hence it is very challenging to detect early stage of Parkinson Disease. In study, work focuses on only confirmed symptoms of Parkinson Disease which doesn’t deals to any other disease completely. Study also focus on the Unified Parkinson’s Disease Rating scale (UPDRS) for Parkinson Disease along with respective symptoms. It includes the analysis in terms of medical science and computer applications for analyzing PD. Medication for PD has also been discussed in the study including wide literature survey which provide the clearance of the goal for treating PD. Proposed rating scale in the study is time efficient as compared with UPDRS.","PeriodicalId":6804,"journal":{"name":"2020 International Conference on Computational Performance Evaluation (ComPE)","volume":"14 1","pages":"121-125"},"PeriodicalIF":0.0,"publicationDate":"2020-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74345719","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-07-01DOI: 10.1109/ComPE49325.2020.9199986
G. Khekare, Shrutika Verma, Samay Sur, R. Haldkar, Pratik Moon
In the agricultural sector, automation being the fundamental property, it increases and improves the standard, broadening, and efficiency of manufacturing goods. The quality of evaluation affected due to improving the sorting of goods. As sorting is one of the most important industry challenges, a reliable sorting method is needed so it became easy to package the goods automatically. Property used in this process is preprocessing, thresholding, segmentation, extraction, classification, and detection. With the help of training and testing images, the machine will identify the goods based on its color, texture, shape, and defects. Thus, this process will lead to a better quality of the image which will help further for packaging the goods in industries. The use of raspberry pi along with the counter sensor and flapper mechanism, the process of automated packaging will improve the quality of results in a better way than before.
{"title":"Design of Fruit Segregation and Packaging Machine","authors":"G. Khekare, Shrutika Verma, Samay Sur, R. Haldkar, Pratik Moon","doi":"10.1109/ComPE49325.2020.9199986","DOIUrl":"https://doi.org/10.1109/ComPE49325.2020.9199986","url":null,"abstract":"In the agricultural sector, automation being the fundamental property, it increases and improves the standard, broadening, and efficiency of manufacturing goods. The quality of evaluation affected due to improving the sorting of goods. As sorting is one of the most important industry challenges, a reliable sorting method is needed so it became easy to package the goods automatically. Property used in this process is preprocessing, thresholding, segmentation, extraction, classification, and detection. With the help of training and testing images, the machine will identify the goods based on its color, texture, shape, and defects. Thus, this process will lead to a better quality of the image which will help further for packaging the goods in industries. The use of raspberry pi along with the counter sensor and flapper mechanism, the process of automated packaging will improve the quality of results in a better way than before.","PeriodicalId":6804,"journal":{"name":"2020 International Conference on Computational Performance Evaluation (ComPE)","volume":"5 1","pages":"709-714"},"PeriodicalIF":0.0,"publicationDate":"2020-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"72784577","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-07-01DOI: 10.1109/ComPE49325.2020.9200076
Rituraj Singh, Jasmeet Singh, M. S. Gill, R. Malhotra, Garima
Software development life cycle comprises of planning, design, implementation, testing and eventually, deployment. Software defect prediction can be used in the initial stages of the development life cycle for identifying defective modules. Researchers have devised various methods that can be used for effective software defect prediction. The prediction of the presence of defects or bugs in a software module can facilitate the testing process as it would enable developers and testers to allocate their time and resources on modules that are prone to defects. Transfer learning can be used for transferring knowledge obtained from one domain into the other. In this paper, we propose Transfer Learning Code Vectorizer, a novel method that derives features from the text of the software source code itself and uses those features for defect prediction. We focus on the software code and convert it into vectors using a pre-trained deep learning language model. These code vectors are subsequently passed through machine and deep learning models. Further, we compare the results of using deep learning on the text of the software code versus the usage of software metrics for prediction of defects. In terms of weighted F1 scores, the experiments show that applying the proposed TLCV method outperforms the other machine learning techniques by 9.052%.
{"title":"Transfer Learning Code Vectorizer based Machine Learning Models for Software Defect Prediction","authors":"Rituraj Singh, Jasmeet Singh, M. S. Gill, R. Malhotra, Garima","doi":"10.1109/ComPE49325.2020.9200076","DOIUrl":"https://doi.org/10.1109/ComPE49325.2020.9200076","url":null,"abstract":"Software development life cycle comprises of planning, design, implementation, testing and eventually, deployment. Software defect prediction can be used in the initial stages of the development life cycle for identifying defective modules. Researchers have devised various methods that can be used for effective software defect prediction. The prediction of the presence of defects or bugs in a software module can facilitate the testing process as it would enable developers and testers to allocate their time and resources on modules that are prone to defects. Transfer learning can be used for transferring knowledge obtained from one domain into the other. In this paper, we propose Transfer Learning Code Vectorizer, a novel method that derives features from the text of the software source code itself and uses those features for defect prediction. We focus on the software code and convert it into vectors using a pre-trained deep learning language model. These code vectors are subsequently passed through machine and deep learning models. Further, we compare the results of using deep learning on the text of the software code versus the usage of software metrics for prediction of defects. In terms of weighted F1 scores, the experiments show that applying the proposed TLCV method outperforms the other machine learning techniques by 9.052%.","PeriodicalId":6804,"journal":{"name":"2020 International Conference on Computational Performance Evaluation (ComPE)","volume":"74 1","pages":"497-502"},"PeriodicalIF":0.0,"publicationDate":"2020-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74141468","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：