scispace - formally typeset
Search or ask a question
Author

K. S. Umadevi

Other affiliations: SRM University
Bio: K. S. Umadevi is an academic researcher from VIT University. The author has contributed to research in topics: Computer science & Wireless sensor network. The author has an hindex of 3, co-authored 25 publications receiving 52 citations. Previous affiliations of K. S. Umadevi include SRM University.

Papers
More filters
Proceedings ArticleDOI
01 Feb 2019
TL;DR: Simulation results indicate Support Vector Machine (SVM) with the TF-IDF gave the most accurate prediction.
Abstract: Fake news consists of news that is not well researched or deliberate steps have been taken to spread misinformation or hoaxes via different forms of news distribution networks. This paper aims to tackle this issue using a computational model of probabilistic and geometric machine learning models. Moreover, the scores of two different vectorizers namely count and Term Frequency Inverse Document Format(TF-IDF) will be compared to find the appropriate vectorizer for fake news detection. English stop words have been used to improve the scores. Various classifiers like Naive Bayes, Support Vector Machine(SVM), Logistic regression and decision tree classifier were used to predict the fake news. Simulation results indicate Support Vector Machine (SVM) with the TF-IDF gave the most accurate prediction.

43 citations

Proceedings ArticleDOI
01 Sep 2018
TL;DR: A stock market prediction model which is considers different parameters of a particular stock and shows that behavior of market can be predicted using machine learning techniques is designed.
Abstract: It has been proved that the decisions in stock market exchange may bring influence on the investors, financial institutions, banking sectors etc. The stock market is a highly composite system in addition often concealed with mystery, it is therefore, very difficult to analyze all the impacting factors before making a decision. In this research, we have tried to design a stock market prediction model which is considers different parameters of a particular stock. Analysis is performed after obtaining the stock scores. This analysis involves visualization of stock scores in the form of various plots and prediction of the scores using a time series model known as ARIMA (auto regressive moving average). The results shows that the time series model performed a descent prediction of the market scores with considerably high accuracy. Each factor was studied independently to find out its association with market performance. Furthermore the results suggests that behavior of market can be predicted using machine learning techniques.

8 citations

Proceedings ArticleDOI
01 Aug 2017
TL;DR: This work study the basic requirements behind the Time Sensitive Network and ingress policies in it, and proposes, the multilevel scheduling mechanism to support various class of service of Ethernet.
Abstract: Time sensitive networking becomes increasingly important in handling control and management information. Utilization of the existing infrastructure is considered to be an important issue due to the increasing demand. Since control/management data packets can use the existing infrastructure, this work study the basic requirements behind the Time Sensitive Network and ingress policies in it. Further proposes, the multilevel scheduling mechanism to support various class of service of Ethernet.

7 citations

Journal ArticleDOI
TL;DR: In this article , the combination of wrapper filtering method using Pearson correlation with recursion function is used to eliminate the unwanted features and then the deep neural network is used for detecting intruders attack over the data in the network.
Abstract: Huge data over the cloud computing and big data are processed over the network. The data may be stored, send, altered and communicated over the network between the source and destination. Once data send by source to destination, before reaching the destination data may be attacked by any intruders over the network. The network has numerous routers and devices to connect to internet. Intruders may attack any were in the network and breaks the original data, secrets. Detection of attack in the network became interesting task for many researchers. There are many intrusion detection feature selection algorithm has been suggested which lags on performance and accuracy. In our article we propose new IDS feature selection algorithm with higher accuracy and performance in detecting the intruders. The combination of wrapper filtering method using Pearson correlation with recursion function is used to eliminate the unwanted features. This feature extraction process clearly extracts the attacked data. Then the deep neural network is used for detecting intruders attack over the data in the network. This hybrid machine learning algorithm in feature extraction process helps to find attacked information using recursive function. Performance of proposed method is compared with existing solution. The traditional feature selection in IDS such as differential equation (DE), Gain ratio (GR), symmetrical uncertainty (SU) and artificial bee colony (ABC) has less accuracy than proposed PCRFE. The experimented results are shown that our proposed PCRFE-CDNN gives 99% of accuracy in IDS feature selection process and 98% in sensitivity.

6 citations

Journal ArticleDOI
TL;DR: In this paper , two cluster models have been constructed and coupled with three Convolutional Neural Networks (CNN) models based on Demand-Side Management (DSM) to determine the optimal operating time, stability, and management model.

6 citations


Cited by
More filters
Journal ArticleDOI
TL;DR: The issues that need to be paid attention to in the research of swarm intelligence algorithm optimization for MWSNs are put forward, and the development trend and prospect of this research direction in the future are prospected.
Abstract: Network performance optimization has always been one of the important research subjects in mobile wireless sensor networks. With the expansion of the application field of MWSNs and the complexity of the working environment, traditional network performance optimization algorithms have become difficult to meet people’s requirements due to their own limitations. The traditional swarm intelligence algorithms have some shortcomings in solving complex practical multi-objective optimization problems. In recent years, scholars have proposed many novel swarm intelligence optimization algorithms, which have strong applicability and achieved good experimental results in solving complex practical problems. These algorithms, like their natural systems of inspiration, show the desirable properties of being adaptive, scalable, and robust. Therefore, the swarm intelligent algorithms (PSO, ACO, ASFA, ABC, SFLA) are widely used in the performance optimization of mobile wireless sensor networks due to its cluster intelligence and biological preference characteristics. In this paper, the main contributions is to comprehensively analyze and summarize the current swarm intelligence optimization algorithm and key technologies of mobile wireless sensor networks, as well as the application of swarm intelligence algorithm in MWSNs. Then, the concept, classification and architecture of Internet of things and MWSNs are described in detail. Meanwhile, the latest research results of the swarm intelligence algorithms in performance optimization of MWSNs are systematically described. The problems and solutions in the performance optimization process of MWSNs are summarized, and the performance of the algorithms in the performance optimization of MWSNs is compared and analyzed. Finally, combined with the current research status in this field, the issues that need to be paid attention to in the research of swarm intelligence algorithm optimization for MWSNs are put forward, and the development trend and prospect of this research direction in the future are prospected.

38 citations

Journal ArticleDOI
TL;DR: The authors survey methods for preprocessing data in natural language, vectorization, dimensionality reduction, machine learning, and quality assessment of information retrieval, and contextualize the identification of fake news, and discuss research initiatives and opportunities.
Abstract: The epidemic spread of fake news is a side effect of the expansion of social networks to circulate news, in contrast to traditional mass media such as newspapers, magazines, radio, and television. Human inefficiency to distinguish between true and false facts exposes fake news as a threat to logical truth, democracy, journalism, and credibility in government institutions. In this paper, we survey methods for preprocessing data in natural language, vectorization, dimensionality reduction, machine learning, and quality assessment of information retrieval. We also contextualize the identification of fake news, and we discuss research initiatives and opportunities.

37 citations

Proceedings ArticleDOI
18 Apr 2020
TL;DR: This study aims to apply natural language processing techniques for text analytics and train deep learning models for detecting fake news based on news title or news content and results are showing that models trained with news content can achieve better performance with computation time being sacrificed.
Abstract: FAKE news has proliferated to a big crowd than before in this digital era, the main factor derives from the rise of social media and direct messaging platform. Techniques of fake news stories detection ingenious, varied, and exciting. This study aims to apply natural language processing (NLP) techniques for text analytics and train deep learning models for detecting fake news based on news title or news content. Solution proposed in this study aims to be applied in real-world social media and eliminate the bad experience for user to receive misleading stories that come from non-reputable source. For NLP techniques, text preprocessing such as regular expression, tokenization, lemmatization and stop words removal are used before vectorizing them into N-gram vectors or sequence vectors using terms frequency inverse document frequency (TF-IDF) or one-hot encoding respectively. Then, TensorFlow is chosen as the framework to be used with built in Keras deep learning libraries that is having a large community and number of commits on Tensorflow GitHub repository that can be enough to build deep learning neural network models. Results from the models are showing that models trained with news content can achieve better performance with computation time being sacrificed while models trained with news title require less computation time to achieve good performance. Also, overall performance of models fed with N-gram vectors are slightly better than models fed with sequence vectors.

35 citations

Proceedings ArticleDOI
20 Jun 2018

28 citations

Journal ArticleDOI
TL;DR: This research proposes the use of composition pattern of web links containing news content as a new source of information for fake news detection and proposes a novel embedding technique, which is called link2vec, an extension of word2vec.
Abstract: Today, the world is under siege from various kinds of fake news ranging from politics to COVID-19. Thus, many scholars have been researching automatic fake news detection based on artificial intelligence and machine learning (AI/ML) to prevent the spread of fake news. The mainstream research on detecting fake news so far has been text-based detection approaches, but they have inherent limitations such as the difficulty of short text processing and language dependency. Thus, as an alternative to the text-based approach, the context-based approach is emerging. The most common context-based approach the use of distributors’ network information in social media. However, such information is difficult to obtain, and only propagation within a single social media can be traced. Under this background, we propose the use of composition pattern of web links containing news content as a new source of information for fake news detection. To properly vectorize the composition pattern of web links, this study proposes a novel embedding technique, which is called link2vec, an extension of word2vec. To test the effectiveness and language independency of our link2vec-based model, we applied it to two real-world fake news datasets in different languages (English and Korean). As comparison models, we adopted the conventional text-based model and a hybrid model that combined text and whitelist-based link information proposed by a prior study. Results revealed that in the datasets in two languages, the link2vec-based detection models outperformed all the comparison models with statistical significance. Our research is expected to contribute to suggesting a completely new path for effective fake news detection.

24 citations