Pub Date : 2020-09-26DOI: 10.1109/IICAIET49801.2020.9257815
Aditi Gupta, Jasy Liew Suet Yan, Cheah Yu-N
Training a classifier for sentiment polarity detection in product reviews when labeled data is not available for a particular domain poses a challenge, which can be addressed through cross-domain sentiment analysis. We experimented with Convolutional Neural Network (CNN) to learn sentiment polarity (positive or negative) from labeled data available in many different source domains and test its performance on a target domain that it is not trained on. Extensive experiments were conducted on 14 different domains using Amazon product reviews. Our preliminary findings show that cross-domain CNN models trained with multiple source domains achieved accuracy of above 80% across all the domains and outperform the in-domain models trained using limited labeled data from the same domain. In fact, the cross-domain CNN models demonstrated better performance when a larger number of source domains are used for training. Therefore, going deep and big is a promising direction to explore for cross-domain sentiment classification.
{"title":"Going Big and Deep: Using Convolutional Neural Network to Leverage Training Data from Multiple Domains for Cross-Domain Sentiment Classification on Product Reviews","authors":"Aditi Gupta, Jasy Liew Suet Yan, Cheah Yu-N","doi":"10.1109/IICAIET49801.2020.9257815","DOIUrl":"https://doi.org/10.1109/IICAIET49801.2020.9257815","url":null,"abstract":"Training a classifier for sentiment polarity detection in product reviews when labeled data is not available for a particular domain poses a challenge, which can be addressed through cross-domain sentiment analysis. We experimented with Convolutional Neural Network (CNN) to learn sentiment polarity (positive or negative) from labeled data available in many different source domains and test its performance on a target domain that it is not trained on. Extensive experiments were conducted on 14 different domains using Amazon product reviews. Our preliminary findings show that cross-domain CNN models trained with multiple source domains achieved accuracy of above 80% across all the domains and outperform the in-domain models trained using limited labeled data from the same domain. In fact, the cross-domain CNN models demonstrated better performance when a larger number of source domains are used for training. Therefore, going deep and big is a promising direction to explore for cross-domain sentiment classification.","PeriodicalId":300885,"journal":{"name":"2020 IEEE 2nd International Conference on Artificial Intelligence in Engineering and Technology (IICAIET)","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130660727","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 : 2020-09-26DOI: 10.1109/IICAIET49801.2020.9257828
A. Ali, Mohd. Yazid Mohd. Anas Khan, N. Bolong, K. A. Maarof, Siti Hasnah Tanalol
Cost-effectiveness in monitoring stormwater quality is challenging in practice, particularly when it involves filtration mechanisms on-site. These challenges arise due to variance in stormwater characteristics, which are lead by rapid urbanization and improper waste management. Hence, an alternative conceptual and design framework of utilizing the concept of IoT (Internet of Things) in monitoring the real-time stormwater quality filtration is discussed. The stormwater quality can be monitored in real-time through data acquisition from wireless network technology in the IoT. ESP32 microcontroller is delegated as the central processing unit for the system. Then, collected data from the sensors of main water quality parameters, including temperature, pH, conductivity, water level, and turbidity, are processed and sent to the webserver while updating the collected data at specified time intervals. It can be remotely accessed via WiFi or GPRS protocol (when WiFi network is not available), regardless of the time and place.
{"title":"Conceptual and design framework for smart stormwater filtration","authors":"A. Ali, Mohd. Yazid Mohd. Anas Khan, N. Bolong, K. A. Maarof, Siti Hasnah Tanalol","doi":"10.1109/IICAIET49801.2020.9257828","DOIUrl":"https://doi.org/10.1109/IICAIET49801.2020.9257828","url":null,"abstract":"Cost-effectiveness in monitoring stormwater quality is challenging in practice, particularly when it involves filtration mechanisms on-site. These challenges arise due to variance in stormwater characteristics, which are lead by rapid urbanization and improper waste management. Hence, an alternative conceptual and design framework of utilizing the concept of IoT (Internet of Things) in monitoring the real-time stormwater quality filtration is discussed. The stormwater quality can be monitored in real-time through data acquisition from wireless network technology in the IoT. ESP32 microcontroller is delegated as the central processing unit for the system. Then, collected data from the sensors of main water quality parameters, including temperature, pH, conductivity, water level, and turbidity, are processed and sent to the webserver while updating the collected data at specified time intervals. It can be remotely accessed via WiFi or GPRS protocol (when WiFi network is not available), regardless of the time and place.","PeriodicalId":300885,"journal":{"name":"2020 IEEE 2nd International Conference on Artificial Intelligence in Engineering and Technology (IICAIET)","volume":"82 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132387957","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 : 2020-09-26DOI: 10.1109/IICAIET49801.2020.9257867
S. Yaacob, Nur Afrina Izzati Affandi, P. Krishnan, Amir Rasyadan, M. Yaakop, Mohamed Fredj
Numerous studies show that driver drowsiness is one of the significant contributors which lead to fatal accidents. Regard to these problems; many hybrid measure detections is proposed using the physiological, behavioural as well as vehicle based. Nevertheless, the proposed model that associates behavioural-based and vehicle-based measure bounce to have a less significant impact on predicting drowsiness as the prediction is based on sensory located closed to the driver. Furthermore, finding drowsiness cannot rely on one single measure of signals. Therefore, this project aimed to produce a hybrid measure detection using multimodal bio signals as it is a gold standard and precisely in evaluating the human body signals. Utilizing the ULg multimodality drowsiness database (called DROZY) database, the electroencephalogram (EEG) and electrocardiogram (ECG) signals have been extracted to determine the drowsiness. k-nearest neighbor (KNN) produces better accuracy than support vector machine (SVM) on both datasets.
{"title":"Drowsiness detection using EEG and ECG signals","authors":"S. Yaacob, Nur Afrina Izzati Affandi, P. Krishnan, Amir Rasyadan, M. Yaakop, Mohamed Fredj","doi":"10.1109/IICAIET49801.2020.9257867","DOIUrl":"https://doi.org/10.1109/IICAIET49801.2020.9257867","url":null,"abstract":"Numerous studies show that driver drowsiness is one of the significant contributors which lead to fatal accidents. Regard to these problems; many hybrid measure detections is proposed using the physiological, behavioural as well as vehicle based. Nevertheless, the proposed model that associates behavioural-based and vehicle-based measure bounce to have a less significant impact on predicting drowsiness as the prediction is based on sensory located closed to the driver. Furthermore, finding drowsiness cannot rely on one single measure of signals. Therefore, this project aimed to produce a hybrid measure detection using multimodal bio signals as it is a gold standard and precisely in evaluating the human body signals. Utilizing the ULg multimodality drowsiness database (called DROZY) database, the electroencephalogram (EEG) and electrocardiogram (ECG) signals have been extracted to determine the drowsiness. k-nearest neighbor (KNN) produces better accuracy than support vector machine (SVM) on both datasets.","PeriodicalId":300885,"journal":{"name":"2020 IEEE 2nd International Conference on Artificial Intelligence in Engineering and Technology (IICAIET)","volume":"118 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116303982","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 : 2020-09-26DOI: 10.1109/iicaiet49801.2020.9257861
{"title":"Copyright","authors":"","doi":"10.1109/iicaiet49801.2020.9257861","DOIUrl":"https://doi.org/10.1109/iicaiet49801.2020.9257861","url":null,"abstract":"","PeriodicalId":300885,"journal":{"name":"2020 IEEE 2nd International Conference on Artificial Intelligence in Engineering and Technology (IICAIET)","volume":"112 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116463437","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 : 2020-09-26DOI: 10.1109/IICAIET49801.2020.9257864
Lu Sin Chong, K. Lim, C. Lee
Stock market prediction has been a challenging task for machine due to time series analysis is needed. In recent years, deep neural networks have been widely applied in many financial time series tasks. Typically, deep neural networks require huge amount of data samples to train a good model. However, the data samples for stock market is limited which caused the networks prone to overfitting. In view of this, this paper leverages deep neural networks with ensemble learning to address this problem. We propose ensemble of Convolutional Neural Network (CNN), Long Short Term Memory (LSTM), and 1DConvNet with LSTM (Conv1DLSTM) to predict the stock market price, named EnsembleDNNs. The performance of the proposed EnsembleDNNs is evaluated with stock market of several companies. The experiment results show encouraging performance as compared to other baselines.
{"title":"Stock Market Prediction using Ensemble of Deep Neural Networks","authors":"Lu Sin Chong, K. Lim, C. Lee","doi":"10.1109/IICAIET49801.2020.9257864","DOIUrl":"https://doi.org/10.1109/IICAIET49801.2020.9257864","url":null,"abstract":"Stock market prediction has been a challenging task for machine due to time series analysis is needed. In recent years, deep neural networks have been widely applied in many financial time series tasks. Typically, deep neural networks require huge amount of data samples to train a good model. However, the data samples for stock market is limited which caused the networks prone to overfitting. In view of this, this paper leverages deep neural networks with ensemble learning to address this problem. We propose ensemble of Convolutional Neural Network (CNN), Long Short Term Memory (LSTM), and 1DConvNet with LSTM (Conv1DLSTM) to predict the stock market price, named EnsembleDNNs. The performance of the proposed EnsembleDNNs is evaluated with stock market of several companies. The experiment results show encouraging performance as compared to other baselines.","PeriodicalId":300885,"journal":{"name":"2020 IEEE 2nd International Conference on Artificial Intelligence in Engineering and Technology (IICAIET)","volume":"124 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114472123","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 : 2020-09-26DOI: 10.1109/IICAIET49801.2020.9257859
Kin Wai Lee, R. Chin
Melanoma skin cancer has been a serious threat due to its high fatality. For this reason, early detection and treatments are given more attention as countermeasures. In recent years, skin cancer detection has been utilizing artificial intelligence techniques, specifically deep convolutional neural network. However, the performance of the convolutional neural network is highly vulnerable to different data constraints, such as the quality and quantity of the data. Therefore, this study explores the synthetization of training data using different data augmentation methods. The work presented in this paper utilizes four different categories of data augmentation methods, which include geometrical transformation, noise addition, colour transformation, and image mix. Multiple layers data augmentation approach is also explored. Dataset expansion strategies and optimized dataset expansion scale are determined to improve the performance of the skin cancer classification. The core findings in our study revealed that single-layer augmentation has better performance than multiple layers augmentation approaches, where region of interest (ROI) image mix has the highest effectiveness compared to other methods. In addition, the best dataset expansion strategy is random ROI image mix. Finally, the optimized dataset expansion is determined at 300%, which yielded the best overall test accuracy at 82.9%, 4.6% improvement compared to unprocessed raw dataset.
{"title":"The Effectiveness of Data Augmentation for Melanoma Skin Cancer Prediction Using Convolutional Neural Networks","authors":"Kin Wai Lee, R. Chin","doi":"10.1109/IICAIET49801.2020.9257859","DOIUrl":"https://doi.org/10.1109/IICAIET49801.2020.9257859","url":null,"abstract":"Melanoma skin cancer has been a serious threat due to its high fatality. For this reason, early detection and treatments are given more attention as countermeasures. In recent years, skin cancer detection has been utilizing artificial intelligence techniques, specifically deep convolutional neural network. However, the performance of the convolutional neural network is highly vulnerable to different data constraints, such as the quality and quantity of the data. Therefore, this study explores the synthetization of training data using different data augmentation methods. The work presented in this paper utilizes four different categories of data augmentation methods, which include geometrical transformation, noise addition, colour transformation, and image mix. Multiple layers data augmentation approach is also explored. Dataset expansion strategies and optimized dataset expansion scale are determined to improve the performance of the skin cancer classification. The core findings in our study revealed that single-layer augmentation has better performance than multiple layers augmentation approaches, where region of interest (ROI) image mix has the highest effectiveness compared to other methods. In addition, the best dataset expansion strategy is random ROI image mix. Finally, the optimized dataset expansion is determined at 300%, which yielded the best overall test accuracy at 82.9%, 4.6% improvement compared to unprocessed raw dataset.","PeriodicalId":300885,"journal":{"name":"2020 IEEE 2nd International Conference on Artificial Intelligence in Engineering and Technology (IICAIET)","volume":"190 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116531050","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 : 2020-09-26DOI: 10.1109/IICAIET49801.2020.9257869
Md. Sazal Miah, Shikder Shafiul Bashar, A. Z. Karim, Zahid Hasan
The purpose of this research is to estimate the heart rate (HR) from wearable gadgets, for example, fingertip gadgets. As the skin of finger-tip is slight, it is not difficult to separate pulse from that point. An optimistic component in this day, HR checking is Photoplethysmography (PPG). Moreover, during physical workout HR extraction precision is truly influenced by clamor and movement artifact (MA). To extract HR variability there are numerous ordinary techniques. In this research, a novel way is utilized to extract HR which is known as a multi-model machine learning technique. In this study, initially training and testing of our developed algorithm is done for various features and various dataset. In addition, separation of noisy and non noisy information is done by K means clustering. Then, the machine gain information from noisy and non noisy dataset. The Linear Regression model is utilized to estimate HR by using dataset. In this study, the feature engineering is also done, as it were, we choose an alternate set of features and know their conduct with our recommended technique and we discover error percentage for each set of features. There were 12 subject from where trial dataset were recorded. The root mean square (RMS) and the mean absolute error of HR was extracted. The lowest absolute mean error we find in this research is 3.06 beats per minute (BPM).
{"title":"Heart Rate Extraction from Photoplethysmography Signal: A Multi Model Machine Learning Approach","authors":"Md. Sazal Miah, Shikder Shafiul Bashar, A. Z. Karim, Zahid Hasan","doi":"10.1109/IICAIET49801.2020.9257869","DOIUrl":"https://doi.org/10.1109/IICAIET49801.2020.9257869","url":null,"abstract":"The purpose of this research is to estimate the heart rate (HR) from wearable gadgets, for example, fingertip gadgets. As the skin of finger-tip is slight, it is not difficult to separate pulse from that point. An optimistic component in this day, HR checking is Photoplethysmography (PPG). Moreover, during physical workout HR extraction precision is truly influenced by clamor and movement artifact (MA). To extract HR variability there are numerous ordinary techniques. In this research, a novel way is utilized to extract HR which is known as a multi-model machine learning technique. In this study, initially training and testing of our developed algorithm is done for various features and various dataset. In addition, separation of noisy and non noisy information is done by K means clustering. Then, the machine gain information from noisy and non noisy dataset. The Linear Regression model is utilized to estimate HR by using dataset. In this study, the feature engineering is also done, as it were, we choose an alternate set of features and know their conduct with our recommended technique and we discover error percentage for each set of features. There were 12 subject from where trial dataset were recorded. The root mean square (RMS) and the mean absolute error of HR was extracted. The lowest absolute mean error we find in this research is 3.06 beats per minute (BPM).","PeriodicalId":300885,"journal":{"name":"2020 IEEE 2nd International Conference on Artificial Intelligence in Engineering and Technology (IICAIET)","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127247107","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 : 2020-09-26DOI: 10.1109/IICAIET49801.2020.9257865
Pooi Shiang Tan, K. Lim, C. Lee, C. Tan
Sound waves are a form of energy produced by a vibrating object that travels through the medium that can be heard. Generally, the sound is used in human communication, music, alert, and so on. Furthermore, it also helps us to understand what are the events that occurring in the moment, and thereby, provide us hints to understand what is happening around us. This has prompt researchers to study on how humans understand what event is occurring based on the sound waves. In recent years, researchers also study on how to equip the machine with this ability, i.e. acoustic event detection. This study focuses on the acoustic event detection which leverage both frequency spectrogram technique and deep learning methods. Initially, a spectrogram image is generated from the acoustic data by using the frequency spectrogram technique. Then, the generated frequency spectrogram is fed into a pre-trained MobileNet model to extract robust features representations. In this work, 1 Dimensional Convolutional Neural Network (1D-CNN) is adopted to train a model for acoustic event detection. The feature representations are extracted from a pre-trained MobileNet. The proposed 1D-CNN consist of several alternatives of convolution and pooling layers. The last pooling layer is flattened and fed into a fully connected layer to classify the events. Dropout is employed to prevent overfitting. The proposed frequency spectrogram with pre-trained MobileNet and 1D-CNN is then evaluated with three datasets, which are Soundscapes1, Soundscapes2, and UrbanSound8k. From the experimental results, the proposed method obtained 81, 86, and 70 F1-score, for Soundscapes1, Soundscapes2, and UrbanSound8k, respectively.
{"title":"Acoustic Event Detection with MobileNet and 1D-Convolutional Neural Network","authors":"Pooi Shiang Tan, K. Lim, C. Lee, C. Tan","doi":"10.1109/IICAIET49801.2020.9257865","DOIUrl":"https://doi.org/10.1109/IICAIET49801.2020.9257865","url":null,"abstract":"Sound waves are a form of energy produced by a vibrating object that travels through the medium that can be heard. Generally, the sound is used in human communication, music, alert, and so on. Furthermore, it also helps us to understand what are the events that occurring in the moment, and thereby, provide us hints to understand what is happening around us. This has prompt researchers to study on how humans understand what event is occurring based on the sound waves. In recent years, researchers also study on how to equip the machine with this ability, i.e. acoustic event detection. This study focuses on the acoustic event detection which leverage both frequency spectrogram technique and deep learning methods. Initially, a spectrogram image is generated from the acoustic data by using the frequency spectrogram technique. Then, the generated frequency spectrogram is fed into a pre-trained MobileNet model to extract robust features representations. In this work, 1 Dimensional Convolutional Neural Network (1D-CNN) is adopted to train a model for acoustic event detection. The feature representations are extracted from a pre-trained MobileNet. The proposed 1D-CNN consist of several alternatives of convolution and pooling layers. The last pooling layer is flattened and fed into a fully connected layer to classify the events. Dropout is employed to prevent overfitting. The proposed frequency spectrogram with pre-trained MobileNet and 1D-CNN is then evaluated with three datasets, which are Soundscapes1, Soundscapes2, and UrbanSound8k. From the experimental results, the proposed method obtained 81, 86, and 70 F1-score, for Soundscapes1, Soundscapes2, and UrbanSound8k, respectively.","PeriodicalId":300885,"journal":{"name":"2020 IEEE 2nd International Conference on Artificial Intelligence in Engineering and Technology (IICAIET)","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122902845","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}
Writing requirements specification documents plays an important role in determining the success of information system development. To compile documents that are consistent, complete and in accordance with standards, both from a technical and business perspective require enough knowledge. Some previous approaches, such as GUI-F framework, propose automated requirements specification document creation with a variety of different methods. However, most of them do not provide detailed guidance on how stakeholders can identify their needs to support the company's business needs. In addition, some methods only focus on documenting high level requirements specification, such as use case diagram. As for the code development process, this only represents very basic information and lack of technical aspects. In our previous work, we proposed a Rule-Based Ontology Framework (ROF) for Auto-Generating Requirements Specification. ROF covers 2 processes in requirements engineering, namely: elicitation and documentation. The output of the elicitation process is a list of final requirements that are stored in an ontology structure, called Requirements Ontology (RO). Using RO, the documentation process automatically generates 2 outputs: process model in the Business Process Model and Notation (BPMN) standard and Software Requirements Specification (SRS) documents in the IEEE standard. The aim of this paper is to conduct a feasibility analysis to prove that ROF is feasible to be implemented in an Information System (IS) projects. ROF is implemented in a case study, an IS project that calculates lecturer workload activity at a university in Indonesia. The feasibility analysis is carried out in stages for each output using qualitative and quantitative methods. The results of the analysis show that that the framework is feasible to be implemented in the IS project to minimize effort in generating requirements specification.
{"title":"Feasibility Analysis of a Rule-Based Ontology Framework (ROF) for Auto-Generation of Requirements Specification","authors":"Amarilis Putri Yanuarifiani, Fang-Fang Chua, Gaik-Yee Chan","doi":"10.1109/IICAIET49801.2020.9257838","DOIUrl":"https://doi.org/10.1109/IICAIET49801.2020.9257838","url":null,"abstract":"Writing requirements specification documents plays an important role in determining the success of information system development. To compile documents that are consistent, complete and in accordance with standards, both from a technical and business perspective require enough knowledge. Some previous approaches, such as GUI-F framework, propose automated requirements specification document creation with a variety of different methods. However, most of them do not provide detailed guidance on how stakeholders can identify their needs to support the company's business needs. In addition, some methods only focus on documenting high level requirements specification, such as use case diagram. As for the code development process, this only represents very basic information and lack of technical aspects. In our previous work, we proposed a Rule-Based Ontology Framework (ROF) for Auto-Generating Requirements Specification. ROF covers 2 processes in requirements engineering, namely: elicitation and documentation. The output of the elicitation process is a list of final requirements that are stored in an ontology structure, called Requirements Ontology (RO). Using RO, the documentation process automatically generates 2 outputs: process model in the Business Process Model and Notation (BPMN) standard and Software Requirements Specification (SRS) documents in the IEEE standard. The aim of this paper is to conduct a feasibility analysis to prove that ROF is feasible to be implemented in an Information System (IS) projects. ROF is implemented in a case study, an IS project that calculates lecturer workload activity at a university in Indonesia. The feasibility analysis is carried out in stages for each output using qualitative and quantitative methods. The results of the analysis show that that the framework is feasible to be implemented in the IS project to minimize effort in generating requirements specification.","PeriodicalId":300885,"journal":{"name":"2020 IEEE 2nd International Conference on Artificial Intelligence in Engineering and Technology (IICAIET)","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115224172","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 : 2020-09-26DOI: 10.1109/IICAIET49801.2020.9257833
S. Yaacob, Nur Iman Zahra Muhamad'Arif, P. Krishnan, Amir Rasyadan, M. Yaakop, Mohamed Fredj
This study aims to provide a solution in determining the drowsiness state among drivers at the early stage. The revolving issue nowadays is the increasing number of traffic crashes due to drowsiness are considerably at an alarming stage. Drowsiness is a state of sleepiness, which leads to the lapse of attention and focuses. Numerous factors caused drowsiness, which can be determined through the biosignals of an individual. A thorough analysis of the bio-signals of drivers, which is the electroencephalogram (EEG), is applied as one of the solutions in handling drowsiness. EEG is significant in measuring drowsiness levels as it shows the electrical activity of the brain. This study analyzes driver behaviour by measuring the brain wave pattern to detect drowsiness. In this study, the brain signals from the subjects were collected using an EEG headset interfaced with the OpenBCI software. The subjective approach, namely, the Karolinska Sleepiness Scale (KSS), is performed to validate the data. This study involves signal processing in examining brain wave patterns by using MATLAB. An alpha frequency band is extracted from the estimation of power spectral density (PSD) using the periodogram method. Classification of all the extracted features by using a decision tree showed high accuracy ranges from 77.1%-97.20% for each of the subjects. Drowsiness managed to be determined based on increasing alpha power.
{"title":"Early driver drowsiness detection using electroencephalography signals","authors":"S. Yaacob, Nur Iman Zahra Muhamad'Arif, P. Krishnan, Amir Rasyadan, M. Yaakop, Mohamed Fredj","doi":"10.1109/IICAIET49801.2020.9257833","DOIUrl":"https://doi.org/10.1109/IICAIET49801.2020.9257833","url":null,"abstract":"This study aims to provide a solution in determining the drowsiness state among drivers at the early stage. The revolving issue nowadays is the increasing number of traffic crashes due to drowsiness are considerably at an alarming stage. Drowsiness is a state of sleepiness, which leads to the lapse of attention and focuses. Numerous factors caused drowsiness, which can be determined through the biosignals of an individual. A thorough analysis of the bio-signals of drivers, which is the electroencephalogram (EEG), is applied as one of the solutions in handling drowsiness. EEG is significant in measuring drowsiness levels as it shows the electrical activity of the brain. This study analyzes driver behaviour by measuring the brain wave pattern to detect drowsiness. In this study, the brain signals from the subjects were collected using an EEG headset interfaced with the OpenBCI software. The subjective approach, namely, the Karolinska Sleepiness Scale (KSS), is performed to validate the data. This study involves signal processing in examining brain wave patterns by using MATLAB. An alpha frequency band is extracted from the estimation of power spectral density (PSD) using the periodogram method. Classification of all the extracted features by using a decision tree showed high accuracy ranges from 77.1%-97.20% for each of the subjects. Drowsiness managed to be determined based on increasing alpha power.","PeriodicalId":300885,"journal":{"name":"2020 IEEE 2nd International Conference on Artificial Intelligence in Engineering and Technology (IICAIET)","volume":"60 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122885005","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：