Pub Date : 2021-09-13DOI: 10.1109/IICAIET51634.2021.9573739
Cassandra Sze Jin Yong, Soon Yee Chong, Chelvam Dasaratha Raman, R. Chin, Sainarayanan Gopalakrishnan, K. Teo
Electrical Impedance Tomography (EIT) estimates the electrical impedance distribution within a medium and produces cross-sectional images of an admittivity distribution inside an electrically conducting object. EIT in biomedicine application was first applied in lung due to it being large organs that allow large conductivity changes and is a promising technique since it allows continuous monitoring of the ventilation distribution. This study aims to explore the potential EIT technique in medical applications, with strategies to enhance the image reconstruction of the lung images. Performance of the enhanced image reconstruction is analyzed through simulation on the thorax Finite Element Model (FEM) based on a thorax CT image generated using NETGEN Mesher. To integrate and simulate EIT image of the thorax model, data are obtained from the forward and inverse model. Graz consensus Reconstruction algorithm for EIT (GREIT) technique is then applied as the consensus linear reconstruction algorithm for lung EIT images. Subsequently, the involvement of 3D imaging opens the opportunity to explore more electrode placement strategies for enhancement in image reconstruction. Performance of the reconstructed images based on electrode numbers and placement strategies are analyzed using the five figures of merits and classified into poor, average and good using Fuzzy Logic (FL). From the analysis, planar-offset configuration with 16-electrodes outperforms all others while planar configuration with 16-electrodes followed closely.
{"title":"Electrical Impedance Tomography with Fuzzy Logic Classification in Lung Image Reconstruction","authors":"Cassandra Sze Jin Yong, Soon Yee Chong, Chelvam Dasaratha Raman, R. Chin, Sainarayanan Gopalakrishnan, K. Teo","doi":"10.1109/IICAIET51634.2021.9573739","DOIUrl":"https://doi.org/10.1109/IICAIET51634.2021.9573739","url":null,"abstract":"Electrical Impedance Tomography (EIT) estimates the electrical impedance distribution within a medium and produces cross-sectional images of an admittivity distribution inside an electrically conducting object. EIT in biomedicine application was first applied in lung due to it being large organs that allow large conductivity changes and is a promising technique since it allows continuous monitoring of the ventilation distribution. This study aims to explore the potential EIT technique in medical applications, with strategies to enhance the image reconstruction of the lung images. Performance of the enhanced image reconstruction is analyzed through simulation on the thorax Finite Element Model (FEM) based on a thorax CT image generated using NETGEN Mesher. To integrate and simulate EIT image of the thorax model, data are obtained from the forward and inverse model. Graz consensus Reconstruction algorithm for EIT (GREIT) technique is then applied as the consensus linear reconstruction algorithm for lung EIT images. Subsequently, the involvement of 3D imaging opens the opportunity to explore more electrode placement strategies for enhancement in image reconstruction. Performance of the reconstructed images based on electrode numbers and placement strategies are analyzed using the five figures of merits and classified into poor, average and good using Fuzzy Logic (FL). From the analysis, planar-offset configuration with 16-electrodes outperforms all others while planar configuration with 16-electrodes followed closely.","PeriodicalId":234229,"journal":{"name":"2021 IEEE International Conference on Artificial Intelligence in Engineering and Technology (IICAIET)","volume":"44 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122896281","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 : 2021-09-13DOI: 10.1109/IICAIET51634.2021.9573950
M. K. Tan, Chun Chong Loo, Kit Guan Lim, Pungut Ibrahim, H. Goh, K. Teo
Renewable energy is gaining more popularity recently. Tidal currents are driven by two different connected bodies trying to equalize their level differences, hence there will be a flow of water from the high-pressure head to the low-pressure head. It is this kind of water flow that makes tidal current suitable for power generation. The main advantage of tidal power is that it can be forecasted easily. Aside from that, sea water has higher density as compared to air, therefore for the same amount of power, the power can be generated at a lower speed. The tidal current model is composed of a permanent magnet synchronous generator, tidal velocity profile, and another two sub-systems. This model is simulated in Matlab. The resultant tidal velocity is made up of 5 different partial tides. The tidal current turbine model is tested with different inputs of pitch angle and tidal current speed. The results show that the maximum generated output power is 295kW when the pitch angle is 2.77°. Furthermore, the higher the tidal current speed, the higher the generated output power. Aside from that, as the pitch angle is gradually increased while keeping the tidal speed constant, the power coefficient will decrease. Maximum Power Point Tracking algorithm which is based on Perturb and Observe (P&O) is used to locate the maximum power coefficient of the system. It can track the maximum power coefficient successfully but there will be oscillation at the steady state. Cuckoo Search via Levy Flight is able to overcome this problem as there will be no oscillation at steady state and this can prevent power loss. The convergence of Cuckoo Search via Levy Flight is two times faster than P&O.
{"title":"Optimized Energy Extraction in Tidal Current Technology using Evolutionary Algorithm","authors":"M. K. Tan, Chun Chong Loo, Kit Guan Lim, Pungut Ibrahim, H. Goh, K. Teo","doi":"10.1109/IICAIET51634.2021.9573950","DOIUrl":"https://doi.org/10.1109/IICAIET51634.2021.9573950","url":null,"abstract":"Renewable energy is gaining more popularity recently. Tidal currents are driven by two different connected bodies trying to equalize their level differences, hence there will be a flow of water from the high-pressure head to the low-pressure head. It is this kind of water flow that makes tidal current suitable for power generation. The main advantage of tidal power is that it can be forecasted easily. Aside from that, sea water has higher density as compared to air, therefore for the same amount of power, the power can be generated at a lower speed. The tidal current model is composed of a permanent magnet synchronous generator, tidal velocity profile, and another two sub-systems. This model is simulated in Matlab. The resultant tidal velocity is made up of 5 different partial tides. The tidal current turbine model is tested with different inputs of pitch angle and tidal current speed. The results show that the maximum generated output power is 295kW when the pitch angle is 2.77°. Furthermore, the higher the tidal current speed, the higher the generated output power. Aside from that, as the pitch angle is gradually increased while keeping the tidal speed constant, the power coefficient will decrease. Maximum Power Point Tracking algorithm which is based on Perturb and Observe (P&O) is used to locate the maximum power coefficient of the system. It can track the maximum power coefficient successfully but there will be oscillation at the steady state. Cuckoo Search via Levy Flight is able to overcome this problem as there will be no oscillation at steady state and this can prevent power loss. The convergence of Cuckoo Search via Levy Flight is two times faster than P&O.","PeriodicalId":234229,"journal":{"name":"2021 IEEE International Conference on Artificial Intelligence in Engineering and Technology (IICAIET)","volume":"32 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125719061","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 : 2021-09-13DOI: 10.1109/IICAIET51634.2021.9573830
Lim Wan Yee, N. A. A. Bakar, N. H. Hassan, N. M. M. Zainuddin, R. Yusoff, N. A. Rahim
Overhang property issue has sustained over the past ten years in Malaysia. Major overhang property issue was contributed from the unsold residential property. Though the government announced to build a data system and provide the housing data to prevent a mismatch of supply-demand in the property market, there are still not many relevant studies or research on predicting residential property prices. Hence, it is essential to understand the factors that influence the price of residential properties. The study aims to predict the price of a residential property by using a machine learning algorithm. Three algorithms were selected, namely Decision Tree, Linear Regression, and Random Forest, tested against the training and testing datasets obtained from the Malaysian Valuation and Property Services Department. Results show that the Random Forest model produced high accuracy with lower r_squared (R2), RMSE, and MAE values. Significantly, the study has contributed a new insight into essential property features that primarily influence the property price, which will be useful for property developers and buyers who wish to invest in the property market.
{"title":"Using Machine Learning to Forecast Residential Property Prices in Overcoming the Property Overhang Issue","authors":"Lim Wan Yee, N. A. A. Bakar, N. H. Hassan, N. M. M. Zainuddin, R. Yusoff, N. A. Rahim","doi":"10.1109/IICAIET51634.2021.9573830","DOIUrl":"https://doi.org/10.1109/IICAIET51634.2021.9573830","url":null,"abstract":"Overhang property issue has sustained over the past ten years in Malaysia. Major overhang property issue was contributed from the unsold residential property. Though the government announced to build a data system and provide the housing data to prevent a mismatch of supply-demand in the property market, there are still not many relevant studies or research on predicting residential property prices. Hence, it is essential to understand the factors that influence the price of residential properties. The study aims to predict the price of a residential property by using a machine learning algorithm. Three algorithms were selected, namely Decision Tree, Linear Regression, and Random Forest, tested against the training and testing datasets obtained from the Malaysian Valuation and Property Services Department. Results show that the Random Forest model produced high accuracy with lower r_squared (R2), RMSE, and MAE values. Significantly, the study has contributed a new insight into essential property features that primarily influence the property price, which will be useful for property developers and buyers who wish to invest in the property market.","PeriodicalId":234229,"journal":{"name":"2021 IEEE International Conference on Artificial Intelligence in Engineering and Technology (IICAIET)","volume":"51 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127447859","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 : 2021-09-13DOI: 10.1109/IICAIET51634.2021.9573654
Sheng Zhu, M. K. Tan, R. Chin, B. Chua, Xiaoxi Hao, K. Teo
Engine failure is one of the major factors caused vehicle breakdown. In the current practice, the engine faults are diagnosed manually by mechanics and the accuracy is highly relied on their experience. Therefore, this study would like to explore the feasibility of implementing auto fault diagnosis using Probabilistic Neural Network (PNN). A benchmarked engine fault model is developed and simulated in Maltab. The proposed algorithm is designed to detect 9 common engine faults based on the information extracted from exhaust gas, such as hydrocarbon (HC), carbon monoxide (CO), oxides of nitrogen (NOx), carbon dioxide (CO2) and dioxygen (O2). The proposed PNN is trained using the collected engine fault data from experiment and the probability density of PNN is determined based on the Parzen window estimation method. Bayes decision rule is implemented for classifying the types of the engine faults. The simulated results show that the proposed algorithm has faster diagnosis speed, higher accuracy and consistent. The algorithm takes 0.038 s in diagnosing the fault and the average accuracy is 98.3 %.
{"title":"Engine Fault Diagnosis using Probabilistic Neural Network","authors":"Sheng Zhu, M. K. Tan, R. Chin, B. Chua, Xiaoxi Hao, K. Teo","doi":"10.1109/IICAIET51634.2021.9573654","DOIUrl":"https://doi.org/10.1109/IICAIET51634.2021.9573654","url":null,"abstract":"Engine failure is one of the major factors caused vehicle breakdown. In the current practice, the engine faults are diagnosed manually by mechanics and the accuracy is highly relied on their experience. Therefore, this study would like to explore the feasibility of implementing auto fault diagnosis using Probabilistic Neural Network (PNN). A benchmarked engine fault model is developed and simulated in Maltab. The proposed algorithm is designed to detect 9 common engine faults based on the information extracted from exhaust gas, such as hydrocarbon (HC), carbon monoxide (CO), oxides of nitrogen (NOx), carbon dioxide (CO2) and dioxygen (O2). The proposed PNN is trained using the collected engine fault data from experiment and the probability density of PNN is determined based on the Parzen window estimation method. Bayes decision rule is implemented for classifying the types of the engine faults. The simulated results show that the proposed algorithm has faster diagnosis speed, higher accuracy and consistent. The algorithm takes 0.038 s in diagnosing the fault and the average accuracy is 98.3 %.","PeriodicalId":234229,"journal":{"name":"2021 IEEE International Conference on Artificial Intelligence in Engineering and Technology (IICAIET)","volume":"178 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132188440","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 : 2021-09-13DOI: 10.1109/IICAIET51634.2021.9573955
Chee-Hong Ting, Y. Leau, Po-Hung Lai, S. Tan, Asni Tahir
Water is the most critical resource in agriculture. However, concerns are raised about low-purity water, which contributes adverse effects to the soil and plant. It causes significant losses to farmers. Hence, this study proposed a project using sensors to identify and predict water and pH levels. Once triggered (water or pH level exceeds or dropped below standard requirement), the sensor can activate the alarm system and notify the target user via email and SMS. In addition, this project includes predicting pH levels by using the data collected from the pH sensor. Raspberry Pi 3 serves as the central processing unit - implementing and powers up the system and enabling sensors to read and display data. This project utilized rapid prototyping, which comprised several phases, which consist of building, testing, and revising until an acceptable prototype is created. Besides, the system is accessed via remot3.it platform, which connects the device to the system. The system interface is displayed through Virtual Network Computing (VNC) viewer. Overall, this study presents the details in developing a gadget capable of displaying water readings and communicating with the target user. Also, the monthly report will be generated and notify the user via email and SMS.
{"title":"Eye-Tank: Monitoring and Predicting Water and pH Level in Smart Farming","authors":"Chee-Hong Ting, Y. Leau, Po-Hung Lai, S. Tan, Asni Tahir","doi":"10.1109/IICAIET51634.2021.9573955","DOIUrl":"https://doi.org/10.1109/IICAIET51634.2021.9573955","url":null,"abstract":"Water is the most critical resource in agriculture. However, concerns are raised about low-purity water, which contributes adverse effects to the soil and plant. It causes significant losses to farmers. Hence, this study proposed a project using sensors to identify and predict water and pH levels. Once triggered (water or pH level exceeds or dropped below standard requirement), the sensor can activate the alarm system and notify the target user via email and SMS. In addition, this project includes predicting pH levels by using the data collected from the pH sensor. Raspberry Pi 3 serves as the central processing unit - implementing and powers up the system and enabling sensors to read and display data. This project utilized rapid prototyping, which comprised several phases, which consist of building, testing, and revising until an acceptable prototype is created. Besides, the system is accessed via remot3.it platform, which connects the device to the system. The system interface is displayed through Virtual Network Computing (VNC) viewer. Overall, this study presents the details in developing a gadget capable of displaying water readings and communicating with the target user. Also, the monthly report will be generated and notify the user via email and SMS.","PeriodicalId":234229,"journal":{"name":"2021 IEEE International Conference on Artificial Intelligence in Engineering and Technology (IICAIET)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114409087","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 : 2021-09-13DOI: 10.1109/IICAIET51634.2021.9573938
L. C. Chin, Marwan Affandi, M. N. Shah, S. Basah, T. Jian, Muhammad Yazid Din
No matter accurate that putting a sensor in its place there is always a possibility that the position of the sensor is not correct. An inaccurate position may produce an error, which eventually affects the result of the measurement. Sensitivity analysis is intended to determine the amount of error that may occur in measurement by varying important parameters slightly in that measurement and calculating the change of the result. In this paper, sensitivity analysis was simulated in the visual tracking system for lower limb joint measurements. In doing the measurements, markers were put on the limbs of the patients at determined positions. Sensitivity analysis was then simulated by moving the points slightly. There was a total of 729 possible positions coming from three marker positions. The effects of the changes for the distances to be measured were analyzed. It is found that the errors depend on the size of the marker; for a 10-mm marker, the maximum error is only 7.85%, which is relatively small for practical application. When the marker diameter is 13 mm, the maximum error is slightly over 10%, which is still acceptable for practical purposes. There are exactly 27 positions that do not produce errors. Knowing these positions will help the user to reduce the error that may occur during the measurement.
{"title":"Sensitivity Analysis of Tracking Point for A Visual Tracking System on Lower Limb Joint Assessment","authors":"L. C. Chin, Marwan Affandi, M. N. Shah, S. Basah, T. Jian, Muhammad Yazid Din","doi":"10.1109/IICAIET51634.2021.9573938","DOIUrl":"https://doi.org/10.1109/IICAIET51634.2021.9573938","url":null,"abstract":"No matter accurate that putting a sensor in its place there is always a possibility that the position of the sensor is not correct. An inaccurate position may produce an error, which eventually affects the result of the measurement. Sensitivity analysis is intended to determine the amount of error that may occur in measurement by varying important parameters slightly in that measurement and calculating the change of the result. In this paper, sensitivity analysis was simulated in the visual tracking system for lower limb joint measurements. In doing the measurements, markers were put on the limbs of the patients at determined positions. Sensitivity analysis was then simulated by moving the points slightly. There was a total of 729 possible positions coming from three marker positions. The effects of the changes for the distances to be measured were analyzed. It is found that the errors depend on the size of the marker; for a 10-mm marker, the maximum error is only 7.85%, which is relatively small for practical application. When the marker diameter is 13 mm, the maximum error is slightly over 10%, which is still acceptable for practical purposes. There are exactly 27 positions that do not produce errors. Knowing these positions will help the user to reduce the error that may occur during the measurement.","PeriodicalId":234229,"journal":{"name":"2021 IEEE International Conference on Artificial Intelligence in Engineering and Technology (IICAIET)","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122216180","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 : 2021-09-13DOI: 10.1109/IICAIET51634.2021.9573901
B. Chan, I. Saad, N. Bolong, Kang Eng Siew
The surface electromyogram was found very useful in muscle activity scanning and diagnosis purposes. With the high demand from the physiotherapist and neurophysiologist, electromyography (EMG) has been developing rapidly to meet the needs. The quantitative analysis of the EMG signal is required to provide particular characteristics of the EMG signal. In this paper, the EMG signals system's design is presented, and the proposed portable EMG system design concept is discussed to improve the current difficulties of EMG signal collection. The sampling frequency of the EMG signal is between 20–500Hz. The EMG signal is received successfully using the wired devices during the contraction of the muscle. The portable non-invasive EMG system was successfully reduce the interference of the signal whereby the movement of the muscle can be easily detected during the data collection.
{"title":"Preliminary Design of Portable Electromyography (EMG) System for Clinical Signal Acquisition","authors":"B. Chan, I. Saad, N. Bolong, Kang Eng Siew","doi":"10.1109/IICAIET51634.2021.9573901","DOIUrl":"https://doi.org/10.1109/IICAIET51634.2021.9573901","url":null,"abstract":"The surface electromyogram was found very useful in muscle activity scanning and diagnosis purposes. With the high demand from the physiotherapist and neurophysiologist, electromyography (EMG) has been developing rapidly to meet the needs. The quantitative analysis of the EMG signal is required to provide particular characteristics of the EMG signal. In this paper, the EMG signals system's design is presented, and the proposed portable EMG system design concept is discussed to improve the current difficulties of EMG signal collection. The sampling frequency of the EMG signal is between 20–500Hz. The EMG signal is received successfully using the wired devices during the contraction of the muscle. The portable non-invasive EMG system was successfully reduce the interference of the signal whereby the movement of the muscle can be easily detected during the data collection.","PeriodicalId":234229,"journal":{"name":"2021 IEEE International Conference on Artificial Intelligence in Engineering and Technology (IICAIET)","volume":"78 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120968405","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 : 2021-09-13DOI: 10.1109/IICAIET51634.2021.9573852
Nadhirah Johari, Mazlina Mamat, A. Chekima
The ability to recognize distress speech is the essence of an intelligent audio surveillance system. With this ability, the surveillance system can be configured to detect specific distress keywords and launch appropriate actions to prevent unwanted incidents from progressing. This paper aims to find potential distress keywords that the audio surveillance system could recognize. The idea is to use a machine learning classifier as the recognition engine. Five distress keywords: ‘Help’, ‘No’, ‘Oi’, ‘Please’, and ‘Tolong’ were selected to be analyzed. A total of 515 audio signals comprising these five distress keywords were collected and used in the training and testing of 27 classifier models, derived from the Decision Tree, Naïve Bias, Support Vector Machine, K-Nearest Neighbour, Ensemble, and Artificial Neural Network. The features extracted from each audio signal are the Mel-frequency Cepstral Coefficients, while the Principal Component Analysis was applied for feature reduction. The results show that the keyword ‘Please’ is the most recognized, followed by ‘Help’, ‘Oi’, ‘No’ and ‘Tolong’, respectively. This observation was achieved using the Ensemble Bagged Trees classifier, which can recognize ‘Please’ with 99% accuracy in training and 100% accuracy in testing.
{"title":"Performance of Machine Learning Classifiers in Distress Keywords Recognition for Audio Surveillance Applications","authors":"Nadhirah Johari, Mazlina Mamat, A. Chekima","doi":"10.1109/IICAIET51634.2021.9573852","DOIUrl":"https://doi.org/10.1109/IICAIET51634.2021.9573852","url":null,"abstract":"The ability to recognize distress speech is the essence of an intelligent audio surveillance system. With this ability, the surveillance system can be configured to detect specific distress keywords and launch appropriate actions to prevent unwanted incidents from progressing. This paper aims to find potential distress keywords that the audio surveillance system could recognize. The idea is to use a machine learning classifier as the recognition engine. Five distress keywords: ‘Help’, ‘No’, ‘Oi’, ‘Please’, and ‘Tolong’ were selected to be analyzed. A total of 515 audio signals comprising these five distress keywords were collected and used in the training and testing of 27 classifier models, derived from the Decision Tree, Naïve Bias, Support Vector Machine, K-Nearest Neighbour, Ensemble, and Artificial Neural Network. The features extracted from each audio signal are the Mel-frequency Cepstral Coefficients, while the Principal Component Analysis was applied for feature reduction. The results show that the keyword ‘Please’ is the most recognized, followed by ‘Help’, ‘Oi’, ‘No’ and ‘Tolong’, respectively. This observation was achieved using the Ensemble Bagged Trees classifier, which can recognize ‘Please’ with 99% accuracy in training and 100% accuracy in testing.","PeriodicalId":234229,"journal":{"name":"2021 IEEE International Conference on Artificial Intelligence in Engineering and Technology (IICAIET)","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133189254","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 : 2021-09-13DOI: 10.1109/IICAIET51634.2021.9573868
Huan Yang Chan, Ramindhran Rajamohan, Gan Keng Hoon, Nur-Hana Samsudin
Ratings and reviews are always the major consideration factor by online course seekers before they join the course. However, it can be time-consuming to read all the information especially the course reviews. In this research work, our objective is to propose a text analytics pipeline that includes text cleaning, text lemmatization, sentiment analysis, text mining, and visualization that can help course seekers to gain a quick insight into the courses as well as enables them to make a quick comparison between multiple courses. The proposed text analytic pipeline was created in Python Jupyter Notebook. Three different Python-related courses were chosen for the study. The proposed text analytics pipeline solution was proved able to achieve our research objective. It can help course seekers to gain a quick insight including the positive and negative reviews into the courses as well as enables them to make a quick comparison between multiple courses. The n-gram analysis and word cloud generated were sufficient to provide an accurate and informative glance into the course. However, it fell short on sentiment analysis especially in detecting the negative reviews.
{"title":"Text Analytics on Course Reviews from Coursera Platform","authors":"Huan Yang Chan, Ramindhran Rajamohan, Gan Keng Hoon, Nur-Hana Samsudin","doi":"10.1109/IICAIET51634.2021.9573868","DOIUrl":"https://doi.org/10.1109/IICAIET51634.2021.9573868","url":null,"abstract":"Ratings and reviews are always the major consideration factor by online course seekers before they join the course. However, it can be time-consuming to read all the information especially the course reviews. In this research work, our objective is to propose a text analytics pipeline that includes text cleaning, text lemmatization, sentiment analysis, text mining, and visualization that can help course seekers to gain a quick insight into the courses as well as enables them to make a quick comparison between multiple courses. The proposed text analytic pipeline was created in Python Jupyter Notebook. Three different Python-related courses were chosen for the study. The proposed text analytics pipeline solution was proved able to achieve our research objective. It can help course seekers to gain a quick insight including the positive and negative reviews into the courses as well as enables them to make a quick comparison between multiple courses. The n-gram analysis and word cloud generated were sufficient to provide an accurate and informative glance into the course. However, it fell short on sentiment analysis especially in detecting the negative reviews.","PeriodicalId":234229,"journal":{"name":"2021 IEEE International Conference on Artificial Intelligence in Engineering and Technology (IICAIET)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131131624","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 : 2021-09-13DOI: 10.1109/IICAIET51634.2021.9573812
Jing Yee Lim, K. Lim, C. Lee
Stock market prediction is a difficult task as it is extremely complex and volatile. Researchers are exploring methods to obtain good performance in stock market prediction. In this paper, we propose a Stacked Bidirectional Long Short-Term Memory (SBLSTM) network for stock market prediction. The proposed SBLSTM stacks three bidirectional LSTM networks to form a deep neural network model that can gain better prediction performance in the stock price forecasting. Unlike LSTM-based methods, the proposed SBLSTM uses bidirectional LSTM layers to obtain the temporal information in both forward and backward directions. In this way, the long-term dependencies from the past and future stock market values are encapsulated. The performance of the proposed SBLSTM is evaluated on six datasets collected from Yahoo Finance. Additionally, the proposed SBLSTM is compared with the state-of-the-art methods using root mean square error. The empirical studies on six datasets demonstrates that the proposed SBLSTM outperforms the state-of-the-art methods.
{"title":"Stacked Bidirectional Long Short-Term Memory for Stock Market Analysis","authors":"Jing Yee Lim, K. Lim, C. Lee","doi":"10.1109/IICAIET51634.2021.9573812","DOIUrl":"https://doi.org/10.1109/IICAIET51634.2021.9573812","url":null,"abstract":"Stock market prediction is a difficult task as it is extremely complex and volatile. Researchers are exploring methods to obtain good performance in stock market prediction. In this paper, we propose a Stacked Bidirectional Long Short-Term Memory (SBLSTM) network for stock market prediction. The proposed SBLSTM stacks three bidirectional LSTM networks to form a deep neural network model that can gain better prediction performance in the stock price forecasting. Unlike LSTM-based methods, the proposed SBLSTM uses bidirectional LSTM layers to obtain the temporal information in both forward and backward directions. In this way, the long-term dependencies from the past and future stock market values are encapsulated. The performance of the proposed SBLSTM is evaluated on six datasets collected from Yahoo Finance. Additionally, the proposed SBLSTM is compared with the state-of-the-art methods using root mean square error. The empirical studies on six datasets demonstrates that the proposed SBLSTM outperforms the state-of-the-art methods.","PeriodicalId":234229,"journal":{"name":"2021 IEEE International Conference on Artificial Intelligence in Engineering and Technology (IICAIET)","volume":"104 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133328635","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}