2022 5th International Conference on Advances in Science and Technology (ICAST)最新文献

{"title":"Computer Navigation using Audio and Video Aid for Amputees and Parkinsons Patients","authors":"Shaurya Magar, Kharanshu Mehta, Arjav Parekh, Abhijit R. Joshi","doi":"10.1109/ICAST55766.2022.10039655","DOIUrl":"https://doi.org/10.1109/ICAST55766.2022.10039655","url":null,"abstract":"A majority of disadvantaged children in underdeveloped countries cannot attend school due to the dearth of proper infrastructure and inadequate reading materials. Due to the pandemic, schools used various technologies to migrate quickly to online classrooms. This has greatly benefitted People with Disabilities since it made learning more accessible by removing physical impediments such as lack of infrastructure. However, ease of computer navigation still poses an issue for the disabled. This paper presents a novel approach to address this issue. We make use of Face Detection and Speech Recognition to simulate certain mouse and keyboard commands. The user can simply move their face to advance the cursor and voice command to perform the desired operation. For that the system uses the in-built camera and microphone, our solution is extremely cost-effective as compared to other commercial solutions.","PeriodicalId":225239,"journal":{"name":"2022 5th International Conference on Advances in Science and Technology (ICAST)","volume":"65 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131654965","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":"Machine learning & deep learning algorithm reviews for diabetic retinopathy detection","authors":"Pranali Hatode, Maniroja M Edinburgh","doi":"10.1109/ICAST55766.2022.10039518","DOIUrl":"https://doi.org/10.1109/ICAST55766.2022.10039518","url":null,"abstract":"An artificial neural network, expressed as (ANN) a very popular algorithm in machine learning, developed from artificial neurons, has less computation capability, and so is less popular amongst the researchers. Deep learning or may also be called as deep neural network (DNN) has evolved with great computation capabilities with graphic processing units and also large amount of data acquisition is possible with DNN. This paper is the pure analysis and review of the performances of the ML and DL algorithms and also compares the parameters of the fundus images as it aims the detection of Diabetic Retinopathy (DR) with better accuracies. It also speaks about the analysis of another type of images particularly Optical Coherence tomography angiography also called as OCTA used for detection of DR and shows how its accuracies of detection and grading of the disease differs from that of the fundus images. DR is an impairment of diabetes has an effect on eyes and is a consequence of damaged retinal blood vessels. If these complications are identified in the very early stage, then the patient's vision misfortune can be avoided.","PeriodicalId":225239,"journal":{"name":"2022 5th International Conference on Advances in Science and Technology (ICAST)","volume":"11 2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131763342","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":"Society Security App","authors":"R. Kanse, Harish Motekar, V. Rathod, S. Ajagekar, D. Kamble, Trupti Patil, Santosh Govind Rathod","doi":"10.1109/ICAST55766.2022.10039537","DOIUrl":"https://doi.org/10.1109/ICAST55766.2022.10039537","url":null,"abstract":"In the housing society, daily life in urban areas has significant things to manage. Our daily necessities, such as water supply, electricity, security, and a variety of other things that are essential in residential life, fall under the purview of housing society management. In the majority of cases, society management practices a traditional method of correspondence. This absolutely has a few limitations and inconveniences. Everyday notices, month-to-month gatherings, comprehensive developments, various contacts for day-to-day needs, security alarms, high-priority communication, and numerous others may not be conveyed as expected in the ongoing situation, as the greater part of the things are getting dealt with physically. To overcome the issues caused by this time- consuming manual framework, an automated framework should be created to decrease human efforts. In most societies, work is done physically. Resident's contact information is noted on the papers. As all work is done on paper, it is undeniably challenging to oversee and monitor basically everything related to work expenses in society. This cooperative society management system will automate the entire daily life of the society. This system likewise keeps the contacts of specialized individuals or organizations that offer various types of assistance, like labour, housekeeping, pest control, equipment maintenance, and so on. There is no automated framework for all of the things that happen in society, so people can understand what's going on. This framework likewise raises invoices against all residents. This process for sustaining a society is designed to address the most widely recognized issues in societies. Utilizing this framework Advance installments or halfway instalments can be effortlessly noted and followed at any level. They require coordination among the society's management and the vendors, which offer some types of assistance, so the appropriate convenience can be given. In this maintenance bill, bills are sent in a simple way","PeriodicalId":225239,"journal":{"name":"2022 5th International Conference on Advances in Science and Technology (ICAST)","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131029565","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":"Microstrip-line fed Dual-band C-shaped Antenna with Monopolar Characteristics for Application in WLAN","authors":"J. Panda, S. Mohapatra, S. K. Dash, A. Bakshi","doi":"10.1109/ICAST55766.2022.10039616","DOIUrl":"https://doi.org/10.1109/ICAST55766.2022.10039616","url":null,"abstract":"In this paper a In this paper, a dual-band planar antenna with monopolar characteristics has been proposed which is fed by a transmission line of microstrip in nature having the C-shaped element of radiation. Two paths of resonance are created in the element of radiation which is a C-shaped structure for the operation in the in the two resonance bands simultaneously at the center frequency of resonance of 2.41 GHz and 5.59 GHz for WLAN application. Two resonating paths are approximately equal to quarter wavelength length at 2.56 GHz ($0.252lambda_{1}$) and 5.49 GHz ($0.185lambda_{2}$). The antenna which is proposed here demonstrate the sufficient gain and pattern of radiation characteristics which which enables the antenna which is proposed suitable for application in WLAMN band of operation","PeriodicalId":225239,"journal":{"name":"2022 5th International Conference on Advances in Science and Technology (ICAST)","volume":"70 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132993384","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":"An Improvised Approach for Website Domain Classification","authors":"A. Dalvi, Binita Soni, Naman Shah, Dishika Mehta, Archi Mamaniya, Swadit Chandan, Smriti Chaudhary","doi":"10.1109/ICAST55766.2022.10039588","DOIUrl":"https://doi.org/10.1109/ICAST55766.2022.10039588","url":null,"abstract":"In today's world, all search engine portals use suggestions to provide users with various options while searching for websites of a genre. The search engine needs to understand the genre of these websites to give the user relevant suggestions for websites. According to a survey, as of June 2021, about 1.86 billion websites are available on the Web. After looking at these statistics, a manual classification seems rather impractical and tedious. Owing to technological advancements, the authors have used machine learning to categorize websites into a group of preset genres provided in a curated dataset prepared in this particular work. The websites are classified into eleven domains according to popularity and user convenience. It was then made accessible through a dynamic web application that used the machine learning model to predict the type of website after taking the user's input. The authors have explored various classification models for this categorization and implemented the most suitable model with the highest accuracy as the final website domain classifier. The proposed model is also verified using the “web genre corpus 2004 (Genre-KI-04)” open dataset created for website genre classification.","PeriodicalId":225239,"journal":{"name":"2022 5th International Conference on Advances in Science and Technology (ICAST)","volume":"2011 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133720989","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":"Fire Detection using HSV Color Picker: A Review","authors":"Aditi Adurkar, Priya Patel, Manasvi Sabnis, Pradnya Patil","doi":"10.1109/ICAST55766.2022.10039624","DOIUrl":"https://doi.org/10.1109/ICAST55766.2022.10039624","url":null,"abstract":"Massive wildfires are destroying homes and habitats around the world. With wildfires, early detection and prevention become keys to containing the fire. Three days ago, France had to evacuate more than 16,000 people threatened by wildfires in the south-west. There were 1,36,604 fire points in India from January 1 to March 31, 2022, according to the Forest Survey of India (FSI). Our project leverages pioneering AI techniques to detect wildfires at very early stages, way faster than the current systems used. However, we believe that these can be automated with the help of machine learning algorithms. The system will seamlessly detect fire with data from various sources, like IoT devices, satellite data, and data provided by users living nearby the forest. The use of different technologies gives us a unique advantage and helps us detect wildfires faster.","PeriodicalId":225239,"journal":{"name":"2022 5th International Conference on Advances in Science and Technology (ICAST)","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115408072","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":"Layer 2 Solutions to Improve the Scalability of Blockchain","authors":"Shweta Shirodkar, Kaustubh Kulkarni, Rushikesh Khanjode, Surekha Kohle, Priyanka Deshmukh, Pradnya Patil","doi":"10.1109/ICAST55766.2022.10039486","DOIUrl":"https://doi.org/10.1109/ICAST55766.2022.10039486","url":null,"abstract":"Blockchain based decentralized currencies are getting much attention these days and being used widely and deployed commercially, many popular platforms are accepting them as a mode of payment. The most popular among all is Bitcoin (BTC) has achieved great success, but these increased transaction volumes introduced problems like low throughput and high transaction processing time. It's not just with the Bitcoin but also other currencies based on Bitcoin which basically use proof of work principle to register a transaction on distributed ledger also facing similar issues leading to question of scalability of blockchain. This paper focuses on the solutions which can be implemented to make the blockchain scalable.","PeriodicalId":225239,"journal":{"name":"2022 5th International Conference on Advances in Science and Technology (ICAST)","volume":"62 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115475179","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":"Sentiment Analysis on Text with Emoticons Using Supervised Algorithm","authors":"Pooja Tikhe, Aditi Chhabria","doi":"10.1109/ICAST55766.2022.10039660","DOIUrl":"https://doi.org/10.1109/ICAST55766.2022.10039660","url":null,"abstract":"Use of Social media is on a high level now a days. Everything going to be socially uploaded with a large interest Usually people uses text as well as emotional icons called emoticons to express their emotions. Sentiment Analysis is to show person's emotions, opinion by their reviews or posts. There are already many techniques to show calculations of sentiment analysis by posts or reviews like naive bayes or support vector machine. These gives better result in sentiment analysis field. Existing system uses unsupervised method of polarity calculation to analyze sentiment from text and emoticons. Without use of predefined dictionary it calculates polarity on the basis of presence of emoticons. However it restrict presence of emoticons to classify text and analyze sentiment. Proposed system works on Supervised learning method. It reduces restriction of presence of emoticons in existing system. It can work on only text, only emoticons as well as combination of both. Preprocessing steps contains feature extraction and removal of stop words. Here c5 classifier has been used for better accuracy and classify posts by rule-set.","PeriodicalId":225239,"journal":{"name":"2022 5th International Conference on Advances in Science and Technology (ICAST)","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116899260","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 CNN-LSTM network to detect Dysarthria using Mel-Frequency Cepstral Coefficients","authors":"K. Vora, Dishant Padalia, Darshil Mehta, Deepak Kumar Sharma","doi":"10.1109/ICAST55766.2022.10039514","DOIUrl":"https://doi.org/10.1109/ICAST55766.2022.10039514","url":null,"abstract":"Dysarthria is a speech problem acquired at birth due to cerebral palsy (CP) or developed after severe brain damage. Dysarthria affects more than 70% of Parkinson's patients and 10% to 65% of people with traumatic brain injury. It is critical to detect dysarthria and other voice speech difficulties early to diagnose the underlying cause. Intelligent systems capable of identifying dysarthria with incredible precision have been developed using audio processing techniques and various deep learning models. This paper presents a hybrid CNN-LSTM model for classifying patients with dysarthria using audio recordings. The CNN-LSTM combination helps capture spatial and temporal information where CNN acts as a feature extractor while LSTM functions as a classifier. The proposed model was trained on the publicly available 9184 audio recordings from the TORGO dataset, and various audio augmentation techniques were employed to generate synthetic data. A total of 128 features were extracted using Mel Frequency Cepstral Coefficients (MFCC) and fed into the architecture as inputs. The K-fold cross-validation technique was used to avoid overfitting and increase the generalization capability of the model. The proposed architecture achieved a state-of-the-art 99.59% accuracy on the dataset. The presented work will minimize the workload of speech pathologists and help them detect dysarthria precisely and effectively.","PeriodicalId":225239,"journal":{"name":"2022 5th International Conference on Advances in Science and Technology (ICAST)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128889205","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":"The Dynamics of a Bug Bounty Platform","authors":"Ashish Bhushan, Vivekananda Billa, Minal Sonkar, Vighnesh Chavan","doi":"10.1109/ICAST55766.2022.10039642","DOIUrl":"https://doi.org/10.1109/ICAST55766.2022.10039642","url":null,"abstract":"A security vulnerability in software, often referred as a bug, is a weakness in software code that arises as a result of unforeseen design decisions or model mathematical inadequacies. It is a security problem that could be used in cyberattacks to gain entry to digitally stored data unless the defect is fixed with a software patch. We found that a growing community of developers can shift the production from inside the company to its ecosystem partners (eg. bug bounty hunters) using a comprehensive framework for code exposure. That is, instead of closed vertical integration, enterprises can opt to innovate utilizing open external contracts. It's preferable if the center of value creation shifts from within the company to outside. Digital commodities, unlike physical items, allow businesses to maximize spillovers. Bug Bounty Programs use crowdsourcing to find bugs. Bug Bounty Programs (BBPs) are becoming standard security practices within enterprises, thanks to the benefits of crowdsourcing flaw and vulnerability management. Bug bounty programs help businesses by enlisting the help of hackers who can find flaws in their software. Because these programs have access to a larger number of hackers or testers, the chances of detecting problems before malicious hackers try to exploit them are higher. It can be a good public relations option for businesses. These programs can also act as a signal to the public and regulators that a company's security program is well-developed. The relevance of these types of programs is expected to grow, as they have become an industry standard that should attract investment from all companies. We evaluate the existing competitors in the market and learn from the case studies accessible while designing our own bug bounty platform. We also look at the Vulnerability Market Mechanism as well as five key aspects of BBP practice: Scoping of BBPs, Timing of Crowd Engagement, Quality of Submission, Researcher-Firm Communication, and Hackers Motivation Management. We examine difficulties in each area and aim to put strategies in place to improve the effectiveness of BBP. This study contributes to research and practice by identifying difficulties and best practices in crowdsourcing information security for rapid vulnerability detection and mitigation.","PeriodicalId":225239,"journal":{"name":"2022 5th International Conference on Advances in Science and Technology (ICAST)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130929468","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}