2022 12th International Conference on Cloud Computing, Data Science & Engineering (Confluence)最新文献

{"title":"Spatio-Temporal distribution characteristic of covid-19 vaccine using time series forecasting","authors":"Raj Talan, S. Rathee, Rashmi Gandhi","doi":"10.1109/Confluence52989.2022.9734172","DOIUrl":"https://doi.org/10.1109/Confluence52989.2022.9734172","url":null,"abstract":"Over 170 nations have been affected from Coronavirus disease 2019(COVID-19). In nearly all the afflicted countries, the number of afflicted people and dying people has been rising at a frightening rate. Our biggest option for halting the pandemic’s spread is a COVID-19 vaccination. But vaccines are an exhaustible resource. Accurate prediction of vaccine distribution by already implemented policies is critical to assisting policymakers in making sufficient decisions in containing COVID-19 pandemic. Forecasting approaches can be utilized, aiding in the development of better plans and the making of sound judgments. These approaches analyze past events to make more accurate predictions about what will happen in the future according to the current implemented strategy. The effectiveness of various LSTM (Long Short-Term Memory) models as well as the ARIMA (Auto-Regressive Integrated Moving Average) model in projecting vaccine distribution for COVID-19 patients.","PeriodicalId":261941,"journal":{"name":"2022 12th International Conference on Cloud Computing, Data Science & Engineering (Confluence)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-01-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131111746","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":"Context-Based Question Answering System with Suggested Questions","authors":"V. Kumari, Srishti Keshari, Yashvardhan Sharma, Lavika Goel","doi":"10.1109/confluence52989.2022.9734207","DOIUrl":"https://doi.org/10.1109/confluence52989.2022.9734207","url":null,"abstract":"Question Answering and Question Generation are well-researched problems in the field of Natural Language Processing and Information Retrieval. This paper aims to demonstrate the use of novel transformer-based models like BERT, AIBERT, and DistilBERT for Question Answering System and the t5 model for Question Generation. The Question Generation task is integrated with the Question Answering System to suggest relevant questions from the input context using the transfer learning-based model. The question generation model generates questions from the context input by the user and uses different models like DistilBERT, RoBERTa for getting answers from the context. Suggested questions are ranked using BM25 scores to show the most relevant question-answer pairs on the top. The input context can be given as PDF or image(extract texts from image).","PeriodicalId":261941,"journal":{"name":"2022 12th International Conference on Cloud Computing, Data Science & Engineering (Confluence)","volume":"233 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-01-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128132185","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":"Modish Energy Datum using LoRa Technology","authors":"C. Kumari, Rohit Kumar, Saurav Gupta, Shourjya Hazra, Mayurakshya Paul","doi":"10.1109/confluence52989.2022.9734228","DOIUrl":"https://doi.org/10.1109/confluence52989.2022.9734228","url":null,"abstract":"The advancement in emerging technologies to make life easier has brought a boon to the world of innovations. In the rural parts of the country, still it has been seen in door-to-door electricity bill collection where a person from the electricity department needs to visit each home to take a household electric meter reading. This manual process of reading bills is a rigid and time taking task. Errors in the reading bill such as extra amount and crime committed in name of reading bills have been recorded in the last few years. In times of pandemic, it became more difficult to read bills, and consumers were pressurized to pay bills of more than 9- 10 months at a single time. To overcome this, there is a need for a smart energy meter that could automate the system of bill reading. This paper introduces the idea of the smart energy meter using LoRa technology. LoRa module which can work with a microprocessor can be very useful to implement this idea. LoRa can transmit a huge amount of data accurately up to a large distance without any significant loss. In big cities, smart energy meter is being used which can read the units and send bills. But since these devices need high-speed network connectivity, hence it is not possible to implement the same technology in rural parts of the country. This paper tends to implement a smart energy meter that could make the existing meter smart, rather than installing a new meter. The implemented idea uses very low energy and works without internet connectivity. The stored data can be accessed by the department globally as it is stored in the cloud.","PeriodicalId":261941,"journal":{"name":"2022 12th International Conference on Cloud Computing, Data Science & Engineering (Confluence)","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-01-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128197352","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":"Smart Crop Recommender System-A Machine Learning Approach","authors":"R. K. Ray, Saneev Kumar Das, S. Chakravarty","doi":"10.1109/Confluence52989.2022.9734173","DOIUrl":"https://doi.org/10.1109/Confluence52989.2022.9734173","url":null,"abstract":"Machine learning has proven its efficacy in solving agricultural problems in the recent years such as crop recommendation, crop yield prediction, and many such. With the advancement in the sub-domain of machine learning i.e., deep learning, multiple problems are minutely solved in agricultural sector. This paper focuses on recommending 22 types of crops with the aid of correlation analysis, distribution analysis, ensembling, and majority voting. A three-tiered framework is proposed in order to implement the crop recommendation problem. It includes data preprocessing, classification, and performance evaluation modules. The feature analysis is done through correlation plots and density distribution followed by classification using ensembling techniques. Finally, performance evaluation is performed using majority voting technique. This article further uses ensembling with base learners i.e., decision trees, random forest, Naïve Bayes, and support vector machines using majority voting. Further, majority voting is used to decide the final performance metrics. The practical visualization of the correlation plot, density-histogram distribution plots, confusion matrices, and performance plot are presented. The accuracy achieved post implementation is 99.54& by using Naïve Bayes classifier. The majority voting ensembler has not shown much accuracy i.e., 98.52&. Thus, Naïve Bayes classifier is proved to be the best fit for this problem statement. Some challenges and future research directions are also epitomized in this article.","PeriodicalId":261941,"journal":{"name":"2022 12th International Conference on Cloud Computing, Data Science & Engineering (Confluence)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-01-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130901637","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 Temporal Convolutional Neural Network Based Activity Recognition Model using a Real-Time Two-Dimensional Single Pose Estimation Framework","authors":"Devansh Srivastav, A. Bajpai, Abhishek Singhal","doi":"10.1109/confluence52989.2022.9734159","DOIUrl":"https://doi.org/10.1109/confluence52989.2022.9734159","url":null,"abstract":"Human pose estimation and human activity recognition are two of the most researched domains in computer vision applications. Surveillance, human-computer interaction and video retrieval, all benefit from robust solutions of this domain. Due to disparities in inter-personal parameters, mobility efficiency, and recording configurations, the process is challenging. In this paper, deep learning algorithms are used for both pose estimation and activity recognition models. The pose estimation model generates an array of objects for each keypoint with their coordinated over a specific interval. This time series data is then consumed by the activity recognition model which was trained using a temporal convolutional neural network. The attained accuracy of the model was found to be 92.7% with a loss of 0.19.","PeriodicalId":261941,"journal":{"name":"2022 12th International Conference on Cloud Computing, Data Science & Engineering (Confluence)","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-01-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132188780","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":"SMS Spam Classification Using Machine Learning Techniques","authors":"Tarun Jain, Payal Garg, Namita Chalil, Aditya Sinha, V. Verma, Rishi Gupta","doi":"10.1109/Confluence52989.2022.9734128","DOIUrl":"https://doi.org/10.1109/Confluence52989.2022.9734128","url":null,"abstract":"Almost every person today owns a mobile phone with at least the most basic facilities like messaging and calling. Spam calls are already infamous for the constant ringing of cell phones for promotional or fraudulent pitching to innocent customers. With the reducing costs of bulk messaging services from network providers, a massive base of these spam calls has shifted to messaging. SMS, standing for a short messaging service, has become a dumping ground of unwanted product descriptions and scam offers. Here, in this scenario, classification becomes a necessity. Classification in this context occurs as separating spam messages from ‘ham’ or legitimate messages. For this aforementioned purpose, we used natural language processing techniques and machine learning algorithms in this paper. We used four simple classification models on a dataset from UCI Machine Learning Repository. We compared the accuracies at the end, which pointed towards the most suitable model being SVM, with an accuracy of 98.797%.","PeriodicalId":261941,"journal":{"name":"2022 12th International Conference on Cloud Computing, Data Science & Engineering (Confluence)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-01-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126077404","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":"Hierarchical Clustering Algorithm Based on Fast and Uniform Segmentation","authors":"Xiaojun Wu, Jingjing Wei, Sheng Yuan, Zihong Chen, Xiaochun Wang","doi":"10.1109/confluence52989.2022.9734143","DOIUrl":"https://doi.org/10.1109/confluence52989.2022.9734143","url":null,"abstract":"Hierarchical clustering algorithm is a very important method in data mining. The disadvantage of hierarchical clustering lies in the time complexity of the algorithm and the one-way irreversibility of the algorithm. The inaccuracy of the conditions for cluster termination is another major disadvantage of hierarchical clustering. Hierarchical clustering requires the final cluster number. But for most datasets, the number of clusters cannot be known in advance. Therefore, a method is proposed to combine the split-based and agglomeration-based hierarchical clustering algorithms to first quickly and uniformly partition the original dataset, and then make similar partitions adaptively merge based on the partition density and partition distance on the basis of these partitions. In this paper, aiming at these defects of hierarchical clustering, a hierarchical clustering algorithm based on fast and uniform segmentation is proposed.","PeriodicalId":261941,"journal":{"name":"2022 12th International Conference on Cloud Computing, Data Science & Engineering (Confluence)","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-01-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126614349","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 comparative study on malaria cell detection using computer vision","authors":"A. Shal, Richa Gupta","doi":"10.1109/Confluence52989.2022.9734136","DOIUrl":"https://doi.org/10.1109/Confluence52989.2022.9734136","url":null,"abstract":"The detection of malaria causing organism is done in labs under a microscope manually by people. In this case, this job being done manually there is a maximum risk of error and false detection can cause a life at stake. so, a fare detection of this disease can help control and cure the disease in time. The traditional way of performing this task includes a lot of manual work to be done by a human which takes a lot of time and efforts to complete it. In order to solve this problem a lot of researchers have proposed different algorithms and model in which they used algorithms and concepts of transfer learning, Deep learning and Computer Vision algorithms like Visual Geometry Group Network (VGG net), Convolution Neural Network, ResNet50, YIQ color space Faster-RCNN and many more to classify and check if the cell image belongs to Uninfected class or Parasitized Class. These approaches were found to be very efficient in terms of accurately classifying an image and fast in terms of time taken to provide results. In order to identify malaria cell though Computer Vision and Deep Neural Network in this paper we have conducted a comparative study among four most efficient and widely used algorithms. These four algorithms will be tested on several performance evaluation parameters like Confusion Matrix, Accuracy, True Positivity Rate and Precision. This will help us to check the different aspects of these algorithms. Also, we will be performing hyperparameter tuning in every algorithm which we use. This will help us to make sure that these algorithms work with their maximum potential. The algorithms used in this paper are Convolution Neural Network, Yolo version 4, Yolo version 5 and Single Shot Detector. In this we will be retraining the whole Single Shot Detector algorithm with our dataset. The reason to choose these algorithms is that they are some of the most popular and widely used algorithms also they are highly efficient, and the main reason is that the core principle on which these algorithms works is different from each other so it will be very useful t compare these algorithms and see how they performs.","PeriodicalId":261941,"journal":{"name":"2022 12th International Conference on Cloud Computing, Data Science & Engineering (Confluence)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-01-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126880674","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":"Bidirectional Encoder Representation from Transformers (BERT) Variants for Procedural Long-Form Answer Extraction","authors":"S. Nitish, R. Darsini, G. Shashank, V. Tejas, Arti Arya","doi":"10.1109/confluence52989.2022.9734142","DOIUrl":"https://doi.org/10.1109/confluence52989.2022.9734142","url":null,"abstract":"Extracting information from large verbose documents is a gruelling task which requires patience and huge amounts of effort. Lengthy documents like Portable Document Formats (PDFs) contain tonnes of information including tables, figures, etc. which makes it hard to retrieve specific pieces of textual content like step-by-step instructions or procedures. To the best of our knowledge, this work is the maiden effort to efficiently extract such long-form procedural answers from the aforementioned information sources. The proposed approach retrieves succinct responses from the relevant PDFs for a given user query using a transformer model embedded with attention mechanism. This model is trained using a self-made dataset namely, Pro-LongQA, consisting of carefully crafted procedure based questions and answers. A comparative study of one of the state-of-the-art transformer models, namely, Bidirectional Encoder Representation from Transformers (BERT) and its different variants such as RoBerta, Albert and DistilBert for the task of long-form question answering is performed. Among which, BERT and RoBerta proved to be the best performing models for this task with an accuracy of 87.2% and 86.4% respectively.","PeriodicalId":261941,"journal":{"name":"2022 12th International Conference on Cloud Computing, Data Science & Engineering (Confluence)","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-01-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115358059","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":"Degree of Accuracy in Credit Card Fraud Detection Using Local Outlier Factor and Isolation Forest Algorithm","authors":"Kanishka Negi, Gaddam Prathik Kumar, G. Raj, S. Sahana, Vishal Jain","doi":"10.1109/confluence52989.2022.9734123","DOIUrl":"https://doi.org/10.1109/confluence52989.2022.9734123","url":null,"abstract":"In this era of digitalization where everyone prefers online-based transactional activities, this increases the demand for a credit card, the fraudulent cases are increasing day by day which causes tremendous loss to an individual. Our model comprises 2 major algorithms and uses anomaly detection as a method to classify fraudulent transactions. With the help of these two algorithms i. e., local outlier factor and Isolation Forest. We are implementing our Machine Learning (ML) Model Credit Card Fraud Detection (CCFD) to get the highest possible degree of accuracy of fraud, these two algorithms in layman rs terms isolate the transaction or it can be considered as an outlier i.e., deviation from a normal and common order which have a high rate of anomaly or fraud transaction.","PeriodicalId":261941,"journal":{"name":"2022 12th International Conference on Cloud Computing, Data Science & Engineering (Confluence)","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-01-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114503617","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}