2022 IEEE World Conference on Applied Intelligence and Computing (AIC)最新文献

{"title":"Deep Learning for Time Series Forecasting – With a focus on Loss Functions and Error Measures","authors":"Sujeeth R Malhathkar, T. S","doi":"10.1109/AIC55036.2022.9848877","DOIUrl":"https://doi.org/10.1109/AIC55036.2022.9848877","url":null,"abstract":"Time Series Forecasting is a significant task in the modern world that finds application in several fields. The use of Deep Learning Models to perform forecasting has gained popularity in the past few years due to their superior performance in comparison to standard statistical models. 13 such models were unique in their approach to forecasting using Deep Learning models in the period 2018-2020 are studied. This work lays special focus on the use of loss function during model training and the use of error measures used to evaluate a model and compare the same with other models. We have observed that although loss functions play a role in the training process, they are not treated as rigorously as other training parameters, and several issues when it comes to reporting model performance using error metrics. The problems are highlighted and suggestions are listed.","PeriodicalId":433590,"journal":{"name":"2022 IEEE World Conference on Applied Intelligence and Computing (AIC)","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-06-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134307794","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}

{"title":"A Wearable Biometric Performance measurement System for Boxing - A SURVEY","authors":"J. Brindha, G. Nallavan","doi":"10.1109/AIC55036.2022.9848915","DOIUrl":"https://doi.org/10.1109/AIC55036.2022.9848915","url":null,"abstract":"The performance measurement during the training period or competition is of major interest in the sports research domain. The requirements for an ideal sports performance measuring sensors are less costly, miniature size, and non-intrusive types. It was made possible by the recent developments in MEMS (Micro-Electro Mechanical Sensors) technology of inertial measurement units. The merging of technology with boxing during training and competition has become more common in today’s sports scenario, which is a motivation for proposing this survey paper. The Inertial sensors are being used more commonly for action recognition and classification, automatic scoring of real-time bouts, head impact and performance monitoring of boxing during training and competition. For combat sports, it has become a trend to use inertial sensor technology. Totally 18 records were considered for survey to display the concept that inertial sensor measurements were primarily used for the measurement of boxing striking/hitting quality, boxing strike classification, automatic scoring of bouts, head impact, machine learning technique used, IMU placement on the body, and validation technique employed in boxing. The methods followed to select and implement the inertial sensor technology appear under-researched, and no appropriate protocols for study the results. Our survey helps in understanding of Inertial Measurement Unit (IMU) into application technology to boxing sport.","PeriodicalId":433590,"journal":{"name":"2022 IEEE World Conference on Applied Intelligence and Computing (AIC)","volume":"68 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-06-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133095577","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}

{"title":"Identification of Tomato Plant Diseases Using CNN- A Comparative Review","authors":"Rishabh Mudgil, Nidhi Garg, Preeti Singh, C. Madhu","doi":"10.1109/AIC55036.2022.9848931","DOIUrl":"https://doi.org/10.1109/AIC55036.2022.9848931","url":null,"abstract":"Background: Tomatoes are extensively farmed as a vegetable or fruit in many regions of the country, and many forms of tomato pests and diseases are met over the whole life cycle of tomatoes; thus, early identification and diagnosis of these diseases is critical. The application of several deep learning approaches for identification of plant diseases has recently gained popularity, with encouraging results obtained for different diseases. Problem: The advancement of architectures based on Convolutional Neural Network (CNN) has considerably improved diagnostic accuracy of tomato diseases by using images of tomato leaves. The integration of CNN in the agricultural sector ensures an increased produce of tomato crop in a sustainable manner. However, complexity and performance time of CNN technique is a significant concern. Objective: In most deep learning models, all characteristics generated at various layers are given equal weighting. Significant characteristics should be learned and transferred to higher layers of the network for more exact classification. To improve the CNN performance, reuse and sharing of images (large dataset) is a great work required for tomato disease detection to get high accuracy. Methods: In this paper, we have presented a comprehensive review of twenty recent and notable works which employ CNN-based detection of tomato plant diseases. A comparative analysis of these works is given, taking accuracy as the metric. Results: A complete assessment of Bacterial Canker and Speck, Yellow Leaf Curl, Brown Rugose Fruit, Crown and Root Rot, Early Blight, Bunchy Top, Stolbur disease identification studies employing CNNs in this study. Conclusion: It is observed that CNN scan identify diseases with high accuracy when enough training data is provided. We anticipate that our study will be a useful resource for agricultural disease researchers employing technology for early disease diagnosis and management.","PeriodicalId":433590,"journal":{"name":"2022 IEEE World Conference on Applied Intelligence and Computing (AIC)","volume":"79 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-06-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132999145","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}

{"title":"Hybrid Feature Approach for Plant Disease Detection and Classification using Machine Learning","authors":"P. Kartikeyan, G. Shrivastava","doi":"10.1109/AIC55036.2022.9848939","DOIUrl":"https://doi.org/10.1109/AIC55036.2022.9848939","url":null,"abstract":"Plant diseases identification and classification is a salient task in the agriculture field and has significant impact on crop quantity and quality. Early detection of plant diseases can contribute to reduce losses and increase crop productivity. Accurate identification and categorization of plant diseases was necessary for enhancing crop cultivation and increased crop production yield, for that an image-processing approach could be used. The proposed hybrid feature extraction technology, which integrates Discrete Wavelet Transform decomposition and Grey Level Co-Occurrence Matrices feature extraction with Support Vector Machine classifier could identify and categorize plant diseases to an extent of 95.16 to 98.38% and gave better performance as compared to another model.","PeriodicalId":433590,"journal":{"name":"2022 IEEE World Conference on Applied Intelligence and Computing (AIC)","volume":"139 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-06-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132431325","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}

{"title":"Blended selection in Ant Colony Optimization for solving Travelling Salesman Problem","authors":"Nidhi Yadav, Probhat Pachung, Vani Agrawal, Jagdish Chand Bansal","doi":"10.1109/AIC55036.2022.9848836","DOIUrl":"https://doi.org/10.1109/AIC55036.2022.9848836","url":null,"abstract":"TSP is one of the most well-known combinatorial optimization problems. Ant Colony optimization is highly recommended to solve discrete optimization problems whereas the selection strategy plays a crucial role in the performance of ACO while solving Travelling Salesman Problem (TSP). There are many selection strategies in ACO to solve TSP, such as roulette wheel selection, ranking selection and annealing selection etc. In ACO, the roulette wheel selection is primarily concerned with exploitation, whereas rank selection is influenced by exploration. Therefore, in this paper, a blend of both roulette wheel and ranking selection is proposed as a new selection strategy in ACO. The proposed selection method is tested over 12 standard TSP instances collected from TSP library TSPLIB. The best results obtained from the above mentioned selection method has been recorded and compared with other three selection methods. The experimental results show that the proposed selection method outperformed with other considered selection methods.","PeriodicalId":433590,"journal":{"name":"2022 IEEE World Conference on Applied Intelligence and Computing (AIC)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-06-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115201237","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}

{"title":"A Cloud Based Novel Framework for Addressing Repetitive Behavior in Autistic Individuals","authors":"B. Alhalabi, Harshal A. Sanghvi, Riki H. Patel, A. Pandya, Elisa Cruz Torres","doi":"10.1109/AIC55036.2022.9848981","DOIUrl":"https://doi.org/10.1109/AIC55036.2022.9848981","url":null,"abstract":"Autism is one of the large-scale challenges amongst the kids in United States. According to the survey, Research shows that one out of each 88 youngsters in the United States has some type of ASD. In the view of survey in 2008, information from the Center of the Disease Control shows a 26 percent increase as compared to 2006 and a 71 percent increase as compared to 2002. Behavioral observations are used to diagnose autism spectrum disorder (ASD). To design a data-driven screening and diagnostic technique for ASD. Analyzing the behavioral patterns in the autistic patients is extremely important. Technological solutions play a huge role in automatically detecting patterns such as Spike in the Autistic kids. In this paper, we propose a Bio-Medical Innovation utilizing remote sensor networks. The major behavioral change which needs to be measured is repetitive motion in the Autistic kids. The paper portrays the framework which will assist the physicians and caretakers of the autistic kids to keep a track of the activities and give them feedback on real time when there is an unusual movement. The major objective of the research is to help the patients with autism to get out of social challenges and compete with the normal people in all occupations. The proposed framework is required to analyze the data by the clinicians and give the reports on real time basis.","PeriodicalId":433590,"journal":{"name":"2022 IEEE World Conference on Applied Intelligence and Computing (AIC)","volume":"142 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-06-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115463089","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}

{"title":"Predicting the Trajectories of the Development of Situations from a Given Phase Based on the Logical Conclusion of the Consequences","authors":"D. Strabykin, A. Krutikov, A. Zemtsov, Alexander Chudinov","doi":"10.1109/AIC55036.2022.9848858","DOIUrl":"https://doi.org/10.1109/AIC55036.2022.9848858","url":null,"abstract":"A method is proposed for predicting the trajectories of situations from a given phase based on the logical conclusion of the consequences. The method makes it possible to solve problems arising when using other forecasting methods related to the complexity of interpretability and justification of the results obtained. The formal description of situations is carried out using first-order predicate calculus formulas, which differ in relative simplicity of interpretation. Forecasting is based on deductive logical deduction of consequences from a knowledge base containing rules for the development of the situation and facts characterizing a given phase of the situation. The method of inference of final consequences with the formation of a description of the logical inference scheme is used. The output scheme represents a long-term forecast, which in general contains several branches that determine different trajectories of the situation. The logical inference scheme clearly represents a long-term forecast and reflects the rationale for the results obtained. It is assumed that the situation develops in discrete time, the report of which begins simultaneously with the adoption of a long-term forecast. As the situation develops, new facts are received that correspond to the events taking place. It is proposed to use a series of short-term forecasts. The short-term fore-cast is made for the next moment in time in accordance with the long-term fore-cast and new facts and determines the events (facts) that are expected at the next moment of the situation development. At the same time, it is taken into account that the incoming facts may not confirm some predicted statements and lead to the exclusion of certain branches of logical inference, thus reducing the number of trajectories of the situation. An example of predicting the trajectories of the development of situations from a given phase based on the logical conclusion of the consequences is given.","PeriodicalId":433590,"journal":{"name":"2022 IEEE World Conference on Applied Intelligence and Computing (AIC)","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-06-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115750716","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}

{"title":"Plants Diseases Prediction Framework: A Image-Based System Using Deep Learning","authors":"Madhu Kirola, Kapil Joshi, S. Chaudhary, Neha Singh, Harishchander Anandaram, Ashulekha Gupta","doi":"10.1109/AIC55036.2022.9848899","DOIUrl":"https://doi.org/10.1109/AIC55036.2022.9848899","url":null,"abstract":"Plant diseases mostly harm the leaves, resulting in a loss in agricultural output’s quality and quantity. Plant disease is the most common cause of large-scale crop mortality. India is a country where people’s livelihoods are heavily reliant on agriculture. The disease has caused chaos in the agricultural industry. The human eye’s perception is not quite as sharp as it needs to be to notice minute variations in the sick leaf region. It needs a complex process that requires both plant expertise and a large amount of processing time. As a result, plant diseases can be detected using machine learning. The disease detection method includes image acquisition, image pre-processing, image segmentation, feature extraction, and classification. To prevent crops at the initial stage from diseases, it is essential to develop an automatic system to diagnose plant diseases and identify its category. The goal of the proposed research is to examine several machine algorithms for plant disease prediction. The paper proposed a framework for disease and healthiness detection in plants and the classification of diseases based on symptoms appearing on a leaf. The diseases are grouped into three categories in the paper: bacterial, viral, and fungal. To conclude, the research paper investigates all of these factors and uses several machine learning(DL) techniques and deep learning(DL) techniques. The machine learning(ML) techniques used in the research work are SVM, KNN, RF(Random Forest), LR (Logistic Regression), and the deep learning(DL) technique used is-Convolutional Neural Network(CNN) for disease prediction in the plants. Following that, a comparison of machine learning and deep learning methodologies was conducted. The RF(Random forest) has the highest accuracy of 97.12 % among machine learning classifiers, however, in comparison to the deep learning model mentioned in the study, the CNN classifier has the highest accuracy of 98.43 %","PeriodicalId":433590,"journal":{"name":"2022 IEEE World Conference on Applied Intelligence and Computing (AIC)","volume":"88 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-06-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128424503","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}

{"title":"COVID-19 Detection Using Raw Chest X-Ray Images","authors":"M. Shukla, A. Rasool, Aditya Jain, Vishwas Sahu, Prerak Verma, A. Dubey","doi":"10.1109/AIC55036.2022.9848872","DOIUrl":"https://doi.org/10.1109/AIC55036.2022.9848872","url":null,"abstract":"COVID-19 has had a lasting effect on the human population around the globe. originating from Wuhan, China, in December 2019, the virus managed to spread worldwide in a short time. Huge waiting time between the detection of symptoms and clinical confirmation of the virus being present in the body has made the virus more fatal; thus, rapid screening of large numbers of suspected patients is essential. Due to inefficiency in pathological testing, alternate ways must be devised to combat these issues. Due to advancements in CAD, integrating radiological images with Artificial Intelligence (AI) can detect the disease accurately. This study proposes a deep learning model for automatic COVID-19 detection using raw Chest X-ray (CXR) images. With 17 convolutional layers, the proposed model is trained to diagnose COVID-19 with an 96.67% accuracy. The model can be used to help the world in numerous ways.","PeriodicalId":433590,"journal":{"name":"2022 IEEE World Conference on Applied Intelligence and Computing (AIC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-06-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128549199","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}

{"title":"A Technology Recommender System Based on Web Crawling and Natural Language Processing","authors":"Hima Bindu Ankem Venkata, Andrea Calazacon, Taha M. Mahmoud, T. Hanne","doi":"10.1109/AIC55036.2022.9848970","DOIUrl":"https://doi.org/10.1109/AIC55036.2022.9848970","url":null,"abstract":"The goal of this study is to develop a prototype for a recommendation system that could assist individuals and organizations without in-depth knowledge of technology products and services to choose an appropriate tool/service that best suits their needs. Recommendation systems, as they are popularly known, are personalized information filtering systems which are usually integrated into various consumer and commercial applications. These personalized systems play a vital role, especially when the user is unsure of what to search for. Our work focuses in particular on web-based recommendation systems for end-users in academia and industries who need recommendations for their software tools and services. This paper focuses on extracting information from the web using Apache Nutch, an open-source web crawler which extracts data from websites widely used for software recommendations. The information extracted is indexed in Elasticsearch, whose Natural Language Processing (NLP) and text mining capabilities are applied to provide appropriate recommendations to the end-users. Kibana dashboards and visualizations are used to visualize the recommendations in a format that is conducive for the end-users.","PeriodicalId":433590,"journal":{"name":"2022 IEEE World Conference on Applied Intelligence and Computing (AIC)","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-06-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127388766","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}