Refine Search

New Search

Result: 7,277,199

(searched for: conference:*)
Page of 727,720
Articles per Page
by
Show export options
  Select all
V. S. Suryaa, Z. Sayf Hussain
Communications in Computer and Information Science pp 500-511; https://doi.org/10.1007/978-3-030-81462-5_45

Abstract:
Covid-19, declared as a pandemic by the World Health Organization (WHO), has infected more than 113 million globally across 221 countries. In this work, we propose a method for automatic detection of coronavirus based on analyzing the Chest X-ray images. The dataset used for the study composes of 1200 Covid-19 infected, 1,345 Viral Pneumonia infected and 1,341 healthy patient X-ray images. We use different CNN architectures pretrained on ImageNet dataset, fine tune them to adapt the dataset and use it as feature extractors. We determine the best feature extractor among them, stack them with fully connected layers and employ different classification approaches such as softmax, XGBoost and Support Vector Machines (SVM). The results show that the stacked CNN model with DenseNet169, fully connected layers and XGBoost achieves an accuracy, recall and F1-score of 99.679% and precision of 99.683%. Hence, the proposed model showcases potential to assist physicians and make the diagnosis process more accurate and efficient.
, W. R. Sam Emmanuel
Communications in Computer and Information Science pp 401-409; https://doi.org/10.1007/978-3-030-81462-5_36

Abstract:
Segmentation of subtle lesions in fundus images has become a vital part of diagnosing ocular diseases such as Diabetic Retinopathy (DR). Diabetic eye disease is characterized by the scattered lesions in the retina. Detection of these lesions at the early stage is important as its progression leads to vision loss if proper treatment is not taken. The main objective of the work is to assist ophthalmologist in the effective diagnosis of eye disease providing timely treatment. This paper focuses on developing a deep learning-based Fusion Network (Fu-Net) with an attention mechanism for lesion segmentation in color fundus images. The network was developed based on the baseline U-Net model with trivial modification in the encoder and decoder part of the model. A multi-feature fusion block (MFuse) is integrated with the encoder of the network to extract the lesion features and a channel attention module is integrated with the decoder part to fuse the feature information effectively. Besides, a modified weighted focal loss function is introduced to mitigate the problem of class imbalance in the fundus image. The computational results obtained signifies the superior performance of the proposed method in the lesion segmentation task.
, Prakhar Yadav
Communications in Computer and Information Science pp 488-499; https://doi.org/10.1007/978-3-030-81462-5_44

Abstract:
It is quite intricate for a buyer to reach the publisher’s advertising slot with many market players in the programmatic era. Auction Duplication, internal deals between Demand & Supply side platforms, and rife fraudulent activities are complicating the existing complex process - leading to a single impression being sold through multiple routes by multiple sellers at multiple prices. The dilemma: Which path should the buyer choose, and what should be the fair price to pay? has been staying put for years. The framework suggested in this paper solves the problem of choosing the best path at the right price in the Video Advertising Landscape, a significant contributor compared to other advertising channels. This framework embraces two techniques named Data Envelopment Analysis, where an unsupervised data set is ranked by estimating the relative efficiencies, and a statistical and machine learning hybrid scoring method based on Classification Modeling to help us decide the path worth bidding. These models’ results are compared with each other to choose the best one based on campaign KPI, i.e., CPM (Cost per 1000 impressions) and VCR (Video Completion rate of the video ad). An average of 6%- 12% reduction in CPM and 1% - 4% increment in VCR is observed across 10 live video ad campaigns. The zenith improvements in CPM reduction give rise to a better return on investment(ROI) than the heuristic approach.
, Niraj Singhal, Mohammad Asim, Ajay Kumar, Mahboob Alam
Communications in Computer and Information Science pp 290-302; https://doi.org/10.1007/978-3-030-81462-5_27

Abstract:
Mobile agents are an emerging computing area which replaces client server computing model. Mobile agents are small piece of code and data works automatically on the behalf of owner. Processed different type of activity during the life cycle of mobile agent and execute code on another host computer. Mobile agents execute their assigned function in malicious heterogeneous environment. Mobile agent is using in many applications like E-commerce, parallel computing, and Network management etc. To provide protection for mobile agents is one of the prime issues in broadens of the mobile agent computing. Today, the greatest challenge to the mobile agent technology is security. The numerous advantages accompanying its usage have been fettered by the security concerns/threats. In this article, propose a new framework for securing the secret key which is based on Shamir’s Secret Share and Error Back propagation ANN. Shamir’s SS has been employed for share generation and BP-ANN is used for secret retrieval. Some keystone features of this scheme are: the shareholders need not disclose their identities at any stage (anonymous), no imposition of threshold limit, liberty to use public communication channel, etc. Furthermore, the ongoing research and developments in ANN technology broadens the scope for improvements in the scheme.
Mansi Mahendru,
Communications in Computer and Information Science pp 512-527; https://doi.org/10.1007/978-3-030-81462-5_46

Abstract:
The emergence of social networks is at a great boom today. Every big news before telecasting on television comes to these forums, therefore raises many dilemmas due to misinterpretation regarding the freedom of speaking. One of this trouble is social intimidation that is very disturbing misbehavior that can cause troubling consequences for the victim. Existing works of social intimidation focuses on only one or two topics of harassment. The main aim of this study is to analyze the hub of social intimidation i.e. twitter, consisting of 25,000 tweets covering five topics of harassment i.e. sexism, racism, appearance related, political and intellectual. Moreover, five machine learning and four deep learning techniques were used namely sequential minimal optimization (SMO), random forest, multinomial naïve bayes, logistic regression (LR), decision tree J48, CNN-CB, CNN-GRU, CNN-LRCN and CNN-Bi-LSTM. Each of the classifiers are evaluated using accuracy, precision, recall and f-measure as a performance metric on the dataset. Results indicate the dominance of CNN-Bi-LSTM and logistic regression among all classifiers used.
Nilima Mohite, Manisha Patil, Anil Gonde, Laxman Waghmare
Communications in Computer and Information Science pp 193-203; https://doi.org/10.1007/978-3-030-81462-5_18

Abstract:
In this paper, Most Significant Bit-plane using Local Ternary Pattern is proposed for retrieval of biomedical images. Biomedical images have been evolved in hospitals and medical institutions for disease diagnosis of patients and to detect and record the patient’s history. Algorithm uses transformation scheme to calculate transformed value of each neighboring pixel present in a local bit plane. The proposed method is generated by calculating two binary pattern from a ternary pattern using difference of local biplane transformed values with the intensity value of center pixel. The retrieval efficiency of proposed method is tested on two standard bio-medical databases OASIS-MRI and MESSIDOR database by using Average Retrieval Precision (ARP). We get 66.17% result on OASIS database and 56.27% on MESSIDOR dataset. The MSBPLTP is evaluated using ARP and compared with existing image retrieval methods.
Jayashree Nair, L. S. Aiswarya, P. R. Sruthy
Communications in Computer and Information Science pp 112-123; https://doi.org/10.1007/978-3-030-81462-5_11

Abstract:
India is the home to a very large number of languages. The Indian languages are rich in literature and has been studied by native and foreign Linguists. Unlike English, Indian languages are Morphologically rich and follows free word-order. Even though there have been efforts towards building morphological analyser for Malayalam and Sanskrit, until now an efficient one is not available. In order to solve this problem we come up with the study on morphological analyzer in Indian languages. Morphological analyser is a linguistic tool that would generate the morphemes of a given word. These rules are based on Indian language linguistics. This paper gives a brief description of the approach used for morphological analyser. With the development of a Python Package that make use of Rule Based Approach for developing Morphological Analyzer. It mainly focusing on noun and this analyzer can be used for Information Retrieval, search engines, Machine Translation, speech recognizer, Text Processing etc.
Yuvraj Anil Jadhav, Sakshi Jitendra Jain, Bhushan Sanjay More, Mayur Sunil Jadhav, Bhushan Chaudhari
Communications in Computer and Information Science pp 124-136; https://doi.org/10.1007/978-3-030-81462-5_12

Abstract:
In few years of development Social Media has been one of the remarkable innovations, which have made communication a lot easier and accessible to all. But, it has some drawbacks associated with it as every other useful appliance has. Cyberbullying is one of its major drawbacks in which a group or individuals are embarrassed or harassed with messages from people. To deal with it, this research firstly surveys the state of social media networks. Then, a system is proposed where cyberbullying will be controlled also there is a knowledge intelligence to gather information about trollers and report them accordingly. The system uses JavaScript and deep learning technology to detect abusive texts and images. This paper proposes the simplest and new technology to spot social media abuse which will help in solving all the cyberbullying issues of all age groups.
Vidit Kumar, Vikas Tripathi, Bhaskar Pant
Communications in Computer and Information Science pp 701-710; https://doi.org/10.1007/978-3-030-81462-5_61

Abstract:
Due to rapid technological advancements, the growth of videos uploaded to the internet has increased exponentially. Most of these videos are free of semantic tags, which makes indexing and retrieval a challenging task, and requires much-needed effective content-based analysis techniques to deal with. On the other hand, supervised representation learning from large-scale labeled dataset demonstrated great success in the image domain. However, creating such a large scale labeled database for videos is expensive and time consuming. To this end, we propose an unsupervised visual representation learning framework, which aims to learn spatiotemporal features by exploiting two pretext tasks i.e. rotation prediction and future frame prediction. The performance of the learned features is analyzed by the nearest neighbor task (video retrieval). For this, we choose the UCF-101 dataset to experiment with. The experimental results shows the competitive performance achieve by our method.
, Nasrin Kabir, Al Amin Biswas,
Communications in Computer and Information Science pp 338-350; https://doi.org/10.1007/978-3-030-81462-5_31

Abstract:
Different forms of sleep disorders have become major health problems among people around the world, and insomnia is one of them. It is a physical condition in which a person faces difficulties to fall asleep at night. It leads to various mental disorders, like anxiety and depression. One of the vital causes of substance abuse is insomnia. This study has proposed a machine learning approach to predict insomnia using different socio-demographic factors of the participants. A multilayer stacking model has been employed in this study to predict the appearance of insomnia in a person. For feature reduction, Principal Component Analysis (PCA) has been used. Our proposed ensemble model has attained an accuracy of 88.60%. The effectiveness of our proposed model has been compared to that of other state-of-the-art ensemble classifiers, like AdaBoost, Gradient Boost, Bagging, and Weighted Voting classifier. The proposed model stated in this study has surpassed the performance of the other ensemble classifiers in terms of different efficacy metrics, like sensitivity, precision, specificity, area under the curve (AUC), accuracy, and F1-score.
Page of 727,720
Articles per Page
by
Show export options
  Select all
Back to Top Top