Pub Date : 2023-04-21DOI: 10.1109/ICAIA57370.2023.10169630
Aniket Aayush, Animesh Khare, Abhijnan Bajpai, M. Aman Kumar, Ramamoorthy Srinath
With the advent of recent technologies, Deep Learning, a subset of machine learning, has gained popularity in solving problems in a variety of domains. One vast field of application for Deep Learning approaches is in the generation of video frames. While Video Interpolation has come a long way, Deep Learning based Video Prediction remains a prominent area of research. We implement a recurrent neural network for the task of video prediction, and then analyze the performance of the model for several different generation ratios and examine the impact of variations in the videos on the model’s ability to stick close to the ground truth. The model is tested on raw videos from Youtube of the ‘Comedian’ category. The quality of the model is evaluated using quantitative and qualitative metrics. The potential use case can be in video streaming to reduce the transmission bandwidth and to generate frames that are not present in the video.
{"title":"Video Prediction using Recurrent Neural Network","authors":"Aniket Aayush, Animesh Khare, Abhijnan Bajpai, M. Aman Kumar, Ramamoorthy Srinath","doi":"10.1109/ICAIA57370.2023.10169630","DOIUrl":"https://doi.org/10.1109/ICAIA57370.2023.10169630","url":null,"abstract":"With the advent of recent technologies, Deep Learning, a subset of machine learning, has gained popularity in solving problems in a variety of domains. One vast field of application for Deep Learning approaches is in the generation of video frames. While Video Interpolation has come a long way, Deep Learning based Video Prediction remains a prominent area of research. We implement a recurrent neural network for the task of video prediction, and then analyze the performance of the model for several different generation ratios and examine the impact of variations in the videos on the model’s ability to stick close to the ground truth. The model is tested on raw videos from Youtube of the ‘Comedian’ category. The quality of the model is evaluated using quantitative and qualitative metrics. The potential use case can be in video streaming to reduce the transmission bandwidth and to generate frames that are not present in the video.","PeriodicalId":196526,"journal":{"name":"2023 International Conference on Artificial Intelligence and Applications (ICAIA) Alliance Technology Conference (ATCON-1)","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126592652","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}
Asthma is a chronic condition that affects the air passages in the lungs, causing symptoms such as cough, wheeze, shortness of breath, and chest tightness, which can be triggered by various factors including viral infections, dust, smoke, pollen, and soaps. It can affect patients’ daily lives in many harsh, debilitating ways, severe cases can lead to emergency health care, hospitalization, and even death. Although asthma can’t be cured, good management with inhaled medications can control the disease and enable people with asthma to lead a normal, active life. One of the ways in which asthma management becomes easier is the prediction of severity of asthma exacerbations in a patient. This model utilizes sensors and data collected from IoT devices and smartphones to predict asthma risk and severity. The model is trained on a dataset of asthma patients and takes into account various factors such as symptoms, triggers, and objective test results. The model is integrated with a non-invasive wearable device through bluetooth. The device itself adopts the latest IoT technologies to collect data about the patient’s whereabouts, their triggers as well as the condition of their disease. As the wearable device collects information from the sensor, this data is stored in the web application, where it can be compared to the previously collected readings to predict the severity of the asthma patient. The web application provides an interface between the patient and the data collected for prediction. This system significantly benefits asthma patients by providing a way to manage their condition better.
{"title":"Noninvasive Wearable Device for Monitoring and Assisting Asthma Patients","authors":"B.M. Himani, Dyuthi Abhitha Prakash, Nandita Mahendra, G.R. Asha","doi":"10.1109/ICAIA57370.2023.10169176","DOIUrl":"https://doi.org/10.1109/ICAIA57370.2023.10169176","url":null,"abstract":"Asthma is a chronic condition that affects the air passages in the lungs, causing symptoms such as cough, wheeze, shortness of breath, and chest tightness, which can be triggered by various factors including viral infections, dust, smoke, pollen, and soaps. It can affect patients’ daily lives in many harsh, debilitating ways, severe cases can lead to emergency health care, hospitalization, and even death. Although asthma can’t be cured, good management with inhaled medications can control the disease and enable people with asthma to lead a normal, active life. One of the ways in which asthma management becomes easier is the prediction of severity of asthma exacerbations in a patient. This model utilizes sensors and data collected from IoT devices and smartphones to predict asthma risk and severity. The model is trained on a dataset of asthma patients and takes into account various factors such as symptoms, triggers, and objective test results. The model is integrated with a non-invasive wearable device through bluetooth. The device itself adopts the latest IoT technologies to collect data about the patient’s whereabouts, their triggers as well as the condition of their disease. As the wearable device collects information from the sensor, this data is stored in the web application, where it can be compared to the previously collected readings to predict the severity of the asthma patient. The web application provides an interface between the patient and the data collected for prediction. This system significantly benefits asthma patients by providing a way to manage their condition better.","PeriodicalId":196526,"journal":{"name":"2023 International Conference on Artificial Intelligence and Applications (ICAIA) Alliance Technology Conference (ATCON-1)","volume":"119 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131012090","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 : 2023-04-21DOI: 10.1109/ICAIA57370.2023.10169179
J. Siddartha Varma, J. Panda, G. Anjaneyulu, S. K. Dash, S. Sahu
A highly miniaturized dual CSRR loaded square patch antenna with asymmetrical slits at diagonal locations is presented in this work. This compact antenna is useful for wireless applications in X band frequency range. This simulated patch antenna has two CSRR unit cells on a square patch to improve axial ratio bandwidth, which is required for circular polarization. It has an impedance bandwidth of 15.35% over the frequency of 10.2 GHz to 11.9 GHz (1700 MHz), which covers the X band frequency range for satellite and wireless applications. It also has circular polarization from 10.55 GHz to 11 GHz with an axial ratio bandwidth percentage of 4.65, which covers the fixed satellite services application. The simulated antenna has an average gain of 5.5 dBi in the operating frequency range.
{"title":"V-Shaped Asymmetric Slit Patch Antenna for Wireless Applications Loaded with Dual Complementary Split Ring Resonators","authors":"J. Siddartha Varma, J. Panda, G. Anjaneyulu, S. K. Dash, S. Sahu","doi":"10.1109/ICAIA57370.2023.10169179","DOIUrl":"https://doi.org/10.1109/ICAIA57370.2023.10169179","url":null,"abstract":"A highly miniaturized dual CSRR loaded square patch antenna with asymmetrical slits at diagonal locations is presented in this work. This compact antenna is useful for wireless applications in X band frequency range. This simulated patch antenna has two CSRR unit cells on a square patch to improve axial ratio bandwidth, which is required for circular polarization. It has an impedance bandwidth of 15.35% over the frequency of 10.2 GHz to 11.9 GHz (1700 MHz), which covers the X band frequency range for satellite and wireless applications. It also has circular polarization from 10.55 GHz to 11 GHz with an axial ratio bandwidth percentage of 4.65, which covers the fixed satellite services application. The simulated antenna has an average gain of 5.5 dBi in the operating frequency range.","PeriodicalId":196526,"journal":{"name":"2023 International Conference on Artificial Intelligence and Applications (ICAIA) Alliance Technology Conference (ATCON-1)","volume":"42 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133778095","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 : 2023-04-21DOI: 10.1109/ICAIA57370.2023.10169528
R. Shobarani, R. Sharmila, M. Kathiravan, A. A. Pandian, Ch Narasimha Chary, K. Vigneshwaran
Melanoma is a type of Skin cancer that spreads rapidly and has a significant death risk if it is not detected early and treated. A prompt and accurate diagnosis can improve the patient’s chances of survival. The creation of a skin cancer diagnostic support system based on computer technologies is highly essential. This study suggests a unique deep transfer learning model for categorizing melanoma malignancy. The proposed system comprises of three main phases including image preprocessing, feature extraction and melanoma classification. The preprocessing phase employs image filters such as mean, median, gaussian and non-local means filter along with histogram equalization techniques to obtain the preprocessed images. Feature extraction and classification are performed using pre-trained Convolutional Neural Network architectures such as DenseNet121, Inception-Resnet-V2 and Xception. Using the ISIC 2020 dataset, the suggested deep learning model’s effectiveness is assessed. The experimental findings show that, in terms of precision and computational expense, the suggested deep transfer learning model performs better than cutting-edge deep learning algorithms.
{"title":"Melanoma Malignancy Prognosis Using Deep Transfer Learning","authors":"R. Shobarani, R. Sharmila, M. Kathiravan, A. A. Pandian, Ch Narasimha Chary, K. Vigneshwaran","doi":"10.1109/ICAIA57370.2023.10169528","DOIUrl":"https://doi.org/10.1109/ICAIA57370.2023.10169528","url":null,"abstract":"Melanoma is a type of Skin cancer that spreads rapidly and has a significant death risk if it is not detected early and treated. A prompt and accurate diagnosis can improve the patient’s chances of survival. The creation of a skin cancer diagnostic support system based on computer technologies is highly essential. This study suggests a unique deep transfer learning model for categorizing melanoma malignancy. The proposed system comprises of three main phases including image preprocessing, feature extraction and melanoma classification. The preprocessing phase employs image filters such as mean, median, gaussian and non-local means filter along with histogram equalization techniques to obtain the preprocessed images. Feature extraction and classification are performed using pre-trained Convolutional Neural Network architectures such as DenseNet121, Inception-Resnet-V2 and Xception. Using the ISIC 2020 dataset, the suggested deep learning model’s effectiveness is assessed. The experimental findings show that, in terms of precision and computational expense, the suggested deep transfer learning model performs better than cutting-edge deep learning algorithms.","PeriodicalId":196526,"journal":{"name":"2023 International Conference on Artificial Intelligence and Applications (ICAIA) Alliance Technology Conference (ATCON-1)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116419055","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 : 2023-04-21DOI: 10.1109/ICAIA57370.2023.10169429
Yasir Yousuf, Javed Dhillon, Sachin Mishra
This For microgrids operating in the islanding mode, the stochastic and intermittent output from renewable energy may result in system frequency deviating from the desired level as renewable energy is increasingly integrated into the power system. This paper discusses a self-tuning based fuzzy PID controller frequency control technique for an island microgrid. The fuel cell, flywheel energy storage system, battery energy storage system, diesel generator, and PV system make up the proposed microgrid system. Due to the growing complexity and nonlinear nature of these systems, the fluctuation in the consumption load and generated power has made frequency regulation difficult. This controller’s main goal is to regulate the frequency of an island microgrid. Self-tuning fuzzy PID controller and PID controller are compared, and the results of the simulation reveal that the fuzzy performs better in terms of frequency deviations.
{"title":"Frequency control of an islanded microgrid using self-tuning fuzzy PID controller","authors":"Yasir Yousuf, Javed Dhillon, Sachin Mishra","doi":"10.1109/ICAIA57370.2023.10169429","DOIUrl":"https://doi.org/10.1109/ICAIA57370.2023.10169429","url":null,"abstract":"This For microgrids operating in the islanding mode, the stochastic and intermittent output from renewable energy may result in system frequency deviating from the desired level as renewable energy is increasingly integrated into the power system. This paper discusses a self-tuning based fuzzy PID controller frequency control technique for an island microgrid. The fuel cell, flywheel energy storage system, battery energy storage system, diesel generator, and PV system make up the proposed microgrid system. Due to the growing complexity and nonlinear nature of these systems, the fluctuation in the consumption load and generated power has made frequency regulation difficult. This controller’s main goal is to regulate the frequency of an island microgrid. Self-tuning fuzzy PID controller and PID controller are compared, and the results of the simulation reveal that the fuzzy performs better in terms of frequency deviations.","PeriodicalId":196526,"journal":{"name":"2023 International Conference on Artificial Intelligence and Applications (ICAIA) Alliance Technology Conference (ATCON-1)","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122178422","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 : 2023-04-21DOI: 10.1109/ICAIA57370.2023.10169730
Joy Almeida, Kushal Shah, Rupali Sawant, Pratima Singh
This research is a study on the impact of a specific stock sentiment based on its news, previous stock movements and finally finding investors sentiment over the stock. This study leverages daily Indian financial news between 2017 and 2021, extracted from various Indian and foreign news sources such as Economic Times, Money Control, Livemint, Business Today, NY Times, WSJ and Washington Post. In this work we propose to analyze news data with a unique pre-processing method that uses vectorization and BERT data processing technology. This is followed by a comparative study and predictive machine learning analysis of following models - Naive Bayes and Recurrent Neural Networks (RNN) with Gated Recurrent Units (GRU), Bi-directional Long Short Term Memory (LSTM) and RNN-LSTM with the pre-processed news data leading us to better accuracy and sentiment findings as compared to other approaches. Based on the comparisons, the results show that - Bi-Directional LSTM layer based on RNN architecture along with BERT Data Processing gives an accuracy of 90.15% leading us to a conclusion of adding a layer of BERT data processing for sentiment analysis to get better results. Further an application feature is being proposed which analyzes real-time stock financial news using RNN-Bi-Directional LSTM, giving a confidence value that is used to calculate overall sentiment of a stock being traded in Indian Stock Exchange for different time frames.
{"title":"Time Frame Analysis for Sentiment Prediction of Stock Based on Financial News using Natural Language Processing","authors":"Joy Almeida, Kushal Shah, Rupali Sawant, Pratima Singh","doi":"10.1109/ICAIA57370.2023.10169730","DOIUrl":"https://doi.org/10.1109/ICAIA57370.2023.10169730","url":null,"abstract":"This research is a study on the impact of a specific stock sentiment based on its news, previous stock movements and finally finding investors sentiment over the stock. This study leverages daily Indian financial news between 2017 and 2021, extracted from various Indian and foreign news sources such as Economic Times, Money Control, Livemint, Business Today, NY Times, WSJ and Washington Post. In this work we propose to analyze news data with a unique pre-processing method that uses vectorization and BERT data processing technology. This is followed by a comparative study and predictive machine learning analysis of following models - Naive Bayes and Recurrent Neural Networks (RNN) with Gated Recurrent Units (GRU), Bi-directional Long Short Term Memory (LSTM) and RNN-LSTM with the pre-processed news data leading us to better accuracy and sentiment findings as compared to other approaches. Based on the comparisons, the results show that - Bi-Directional LSTM layer based on RNN architecture along with BERT Data Processing gives an accuracy of 90.15% leading us to a conclusion of adding a layer of BERT data processing for sentiment analysis to get better results. Further an application feature is being proposed which analyzes real-time stock financial news using RNN-Bi-Directional LSTM, giving a confidence value that is used to calculate overall sentiment of a stock being traded in Indian Stock Exchange for different time frames.","PeriodicalId":196526,"journal":{"name":"2023 International Conference on Artificial Intelligence and Applications (ICAIA) Alliance Technology Conference (ATCON-1)","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127176295","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 : 2023-04-21DOI: 10.1109/ICAIA57370.2023.10169516
Jahnavi Gurrala, Rama Vamsi Swarna, S. Panda
The present paper discusses the development of a thumbprint-based financial security scheme for bank lockers that transmits photos of the person opening the locker. This may be accomplished in homes, workplaces, and banks. Everyone who accesses the bank locker should be concerned about security. We frequently forget or lose the key to our bank locker. The bank locker becomes extremely challenging to unlock in these circumstances. It prompts a completely autonomous system for detecting and managing locker rooms at banks. Our security system frequently picks up on prohibited access in the locker room area during robberies. When there is a robbery with a lack of evidence, the banks are unable to identify the thief using the prevalent human-operated security system. To ensure the utmost security of the bank locker room, we proposed an advanced system that identifies as well as restricts those who are attempting to access lockers illegally. The classic locker system that employs keys has been improved with the fingerprint-based locker system. Additionally, the keys need to be protected and might disappear if carelessness is present. results in the proposed system emphasizes an added security to the thumb-print system by including the automatic capture and sending of photograph. The work done here complements the efforts for improving security in banklocker system.
{"title":"Thumbprint-Based Financial Locker Framework using IOT","authors":"Jahnavi Gurrala, Rama Vamsi Swarna, S. Panda","doi":"10.1109/ICAIA57370.2023.10169516","DOIUrl":"https://doi.org/10.1109/ICAIA57370.2023.10169516","url":null,"abstract":"The present paper discusses the development of a thumbprint-based financial security scheme for bank lockers that transmits photos of the person opening the locker. This may be accomplished in homes, workplaces, and banks. Everyone who accesses the bank locker should be concerned about security. We frequently forget or lose the key to our bank locker. The bank locker becomes extremely challenging to unlock in these circumstances. It prompts a completely autonomous system for detecting and managing locker rooms at banks. Our security system frequently picks up on prohibited access in the locker room area during robberies. When there is a robbery with a lack of evidence, the banks are unable to identify the thief using the prevalent human-operated security system. To ensure the utmost security of the bank locker room, we proposed an advanced system that identifies as well as restricts those who are attempting to access lockers illegally. The classic locker system that employs keys has been improved with the fingerprint-based locker system. Additionally, the keys need to be protected and might disappear if carelessness is present. results in the proposed system emphasizes an added security to the thumb-print system by including the automatic capture and sending of photograph. The work done here complements the efforts for improving security in banklocker system.","PeriodicalId":196526,"journal":{"name":"2023 International Conference on Artificial Intelligence and Applications (ICAIA) Alliance Technology Conference (ATCON-1)","volume":"45 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127475102","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 : 2023-04-21DOI: 10.1109/ICAIA57370.2023.10169765
J. Verma, A. Bhandari, Gurpreet Singh
The proliferation of cloud computing is directly responsible for the current transformation phase that the information technology sector is going through. The concept of cloud computing is still in its infancy, yet it is altering the information technology industry. Due to the distributed and open nature of cloud services, they are vulnerable to various threats, including malicious activities and intrusions. Cloud services are also prone to be hacked. Conventional network intrusion detection systems (NIDS) are ineffective against today’s high-volume network traffic because they are trained using a single dataset. The infrastructure and application pose limitations, making processing enormous network traffic in real-time challenging. To protect the cloud from the numerous cloud-based dangers that exist, it is essential to embody Network intrusion detection systems (NIDS) which are equipped with intelligence. This research presents a solution to a modern problem: the development of a distributed and sophisticated NIDS framework using cloud-based solutions. An intelligent NIDS for cloud platforms is proposed in this article, along with an orchestration of a Docker-based Spark cluster over Kubernetes, which is hosted on AWS EC2 instances. The ANN-based NIDS that has been proposed attains an accuracy of 96.3% and encourages Precision scores of 97.2%, Recall scores of 97.5%, and F1-scores of 97.3%.
{"title":"Spark-based Distributed Intelligent Network Intrusion Detection System for Unified Dataset","authors":"J. Verma, A. Bhandari, Gurpreet Singh","doi":"10.1109/ICAIA57370.2023.10169765","DOIUrl":"https://doi.org/10.1109/ICAIA57370.2023.10169765","url":null,"abstract":"The proliferation of cloud computing is directly responsible for the current transformation phase that the information technology sector is going through. The concept of cloud computing is still in its infancy, yet it is altering the information technology industry. Due to the distributed and open nature of cloud services, they are vulnerable to various threats, including malicious activities and intrusions. Cloud services are also prone to be hacked. Conventional network intrusion detection systems (NIDS) are ineffective against today’s high-volume network traffic because they are trained using a single dataset. The infrastructure and application pose limitations, making processing enormous network traffic in real-time challenging. To protect the cloud from the numerous cloud-based dangers that exist, it is essential to embody Network intrusion detection systems (NIDS) which are equipped with intelligence. This research presents a solution to a modern problem: the development of a distributed and sophisticated NIDS framework using cloud-based solutions. An intelligent NIDS for cloud platforms is proposed in this article, along with an orchestration of a Docker-based Spark cluster over Kubernetes, which is hosted on AWS EC2 instances. The ANN-based NIDS that has been proposed attains an accuracy of 96.3% and encourages Precision scores of 97.2%, Recall scores of 97.5%, and F1-scores of 97.3%.","PeriodicalId":196526,"journal":{"name":"2023 International Conference on Artificial Intelligence and Applications (ICAIA) Alliance Technology Conference (ATCON-1)","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130003464","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 : 2023-04-21DOI: 10.1109/ICAIA57370.2023.10169793
Afifi Chaima, Khebizi Ali, Halimi Khaled
In recent years, there has been an intense interest in extracting knowledge from Business Process (BP) execution data provided by Information System (IS). In this area, a set of Process Mining (PM) approaches has been developed. While such conventional PM approaches aim to extract hand-crafted features from the event log, the Deep Learning (DL) models are used to automatically extract the features from the input data. Whereas, the graph representation is the advanced and powerful input format for these DL models. This paper focuses on the pre-processing data representation stage as a starting step for the application of any Machine Learning (ML) technique (process discovery, anomaly detection, classification, recommendation, $ldots$etc.). This phase aim to represent the BP event-log data transitions as Behavior Graphs (BG). This BG constitutes the backbone of our perspective hierarchical DL framework’s based feature extraction and which allows to learn the unified execution of the process hidden behind the event-log data trace’s.
{"title":"Extracting and Exploiting the Behavior Business Process Graph through Transition-Centric Event-Log data","authors":"Afifi Chaima, Khebizi Ali, Halimi Khaled","doi":"10.1109/ICAIA57370.2023.10169793","DOIUrl":"https://doi.org/10.1109/ICAIA57370.2023.10169793","url":null,"abstract":"In recent years, there has been an intense interest in extracting knowledge from Business Process (BP) execution data provided by Information System (IS). In this area, a set of Process Mining (PM) approaches has been developed. While such conventional PM approaches aim to extract hand-crafted features from the event log, the Deep Learning (DL) models are used to automatically extract the features from the input data. Whereas, the graph representation is the advanced and powerful input format for these DL models. This paper focuses on the pre-processing data representation stage as a starting step for the application of any Machine Learning (ML) technique (process discovery, anomaly detection, classification, recommendation, $ldots$etc.). This phase aim to represent the BP event-log data transitions as Behavior Graphs (BG). This BG constitutes the backbone of our perspective hierarchical DL framework’s based feature extraction and which allows to learn the unified execution of the process hidden behind the event-log data trace’s.","PeriodicalId":196526,"journal":{"name":"2023 International Conference on Artificial Intelligence and Applications (ICAIA) Alliance Technology Conference (ATCON-1)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126393387","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 : 2023-04-21DOI: 10.1109/ICAIA57370.2023.10169711
Yuji Sasaki, Keito Tanemura, Yuki Tokuni, R. Miyadera, Hikaru Manabe
This study proposes a method for solving unsolved mathematical games using symbolic regression libraries. We aimed to demonstrate the effectiveness of genetic programming in mathematics in rendering the process of finding formulas more efficient. In the first part of the study, we customized the Python symbolic regression library “gplearn” by adding new features, such as conditional branching. The library uses genetic programming to obtain formulas from data, and we found that the performance of the customized version was better than that of the original. However, the user of this library must be experienced in mathematics to set the conditions for branching. The second part of the study involved the creation of a Swift symbolic regression library using genetic programming. We implemented a new method that combines two criteria for selecting the best formulas: the mean absolute error and the percentage of data described by the formula without error. This new library can discover formulas as good as those discovered using the customized “gplearn” library without requiring specialized knowledge. In some cases, the Swift library discovered formulas that better described the data better than the “gplearn” library.The results of this study suggest the potential for using genetic programming in mathematics and expanding the scope of research on symbolic regression.
{"title":"Application of Symbolic Regression to Unsolved Mathematical Problems","authors":"Yuji Sasaki, Keito Tanemura, Yuki Tokuni, R. Miyadera, Hikaru Manabe","doi":"10.1109/ICAIA57370.2023.10169711","DOIUrl":"https://doi.org/10.1109/ICAIA57370.2023.10169711","url":null,"abstract":"This study proposes a method for solving unsolved mathematical games using symbolic regression libraries. We aimed to demonstrate the effectiveness of genetic programming in mathematics in rendering the process of finding formulas more efficient. In the first part of the study, we customized the Python symbolic regression library “gplearn” by adding new features, such as conditional branching. The library uses genetic programming to obtain formulas from data, and we found that the performance of the customized version was better than that of the original. However, the user of this library must be experienced in mathematics to set the conditions for branching. The second part of the study involved the creation of a Swift symbolic regression library using genetic programming. We implemented a new method that combines two criteria for selecting the best formulas: the mean absolute error and the percentage of data described by the formula without error. This new library can discover formulas as good as those discovered using the customized “gplearn” library without requiring specialized knowledge. In some cases, the Swift library discovered formulas that better described the data better than the “gplearn” library.The results of this study suggest the potential for using genetic programming in mathematics and expanding the scope of research on symbolic regression.","PeriodicalId":196526,"journal":{"name":"2023 International Conference on Artificial Intelligence and Applications (ICAIA) Alliance Technology Conference (ATCON-1)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121036686","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：