{"title":"Use of Machine Learning Techniques to Predict Diabetes at an Early Stage","authors":"Sarra Samet, Mohamed Ridda Laouar, Issam Bendib","doi":"10.1109/icnas53565.2021.9628903","DOIUrl":"https://doi.org/10.1109/icnas53565.2021.9628903","url":null,"abstract":"Because of its ability to improve disease prediction, machine learning has taken a prominent place in healthcare services (HCS). Artificial intelligence and machine learning techniques have already been used in this area. In order to anticipate disease at an early stage, data mining techniques are commonly used. Diabetes has recently become a well-known public chronic condition all across the world. It is rapidly increasing as a result of improper lifestyles, increased consumption of junk food, and a lack of health awareness. Predictive analytics in healthcare is a difficult task, but it can ultimately assist practitioners in making timely decisions regarding a patient’s health and treatment based on huge data. For the purpose of predicting diabetes, seven of the most important machine learning classification techniques have been examined. As a result of a comparison of the multiple machine learning approaches utilized in this study, it has been determined which algorithm is best for prediction of diabetes. With an F1 score of 0,94, XGBoost outperformed other classifiers. To help doctors and practitioners anticipate diabetes earlier using machine learning approaches with more accuracy, this study was written. Models were shown to be more effective than existing work.","PeriodicalId":321454,"journal":{"name":"2021 International Conference on Networking and Advanced Systems (ICNAS)","volume":"76 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127286806","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 New Steganography Technique Based on Dotted Arabic Letters Features","authors":"Abdennour Boulesnane, Adel Beggag, Mustapha Zedadik","doi":"10.1109/icnas53565.2021.9628914","DOIUrl":"https://doi.org/10.1109/icnas53565.2021.9628914","url":null,"abstract":"Because of the great evolution of communication technologies, the subject of information security has become a sensitive topic that must be given great importance in order to protect confidential information. Steganography is one of the effective ways to improve security and information protection while transferring data over the Internet. It consists of hiding information within an unremarkable cover object that can be either of text, image, audio, or video. Arabic language and other similar languages such as Urdu and Persian, possess some special features that make them excellent covers for steganography. In this work, we propose to use for the first time in the literature, the position of points in dotted Arabic letters combining with other Arabic text features to create a new steganography technique. Based on this idea, we introduce new secret bit patterns that could allow us to hide three bits of secret information into a cover object instead of one bit, and hence greatly enhance the capacity performance. Experiments show the effectiveness of our proposed technique by outperforming other existing steganography methods.","PeriodicalId":321454,"journal":{"name":"2021 International Conference on Networking and Advanced Systems (ICNAS)","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121625786","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}