Browse Items (2150 total)
Sort by:
-
Fake News Detection: An Effective Content-Based Approach Using Machine Learning Techniques
Fake news is any information fabricated to mislead readers to spread an idea for certain gains (usually political or financial). In today's world, accessing and sharing information is very fast and almost free. Internet users are growing significantly than ever before. Therefore, online platforms are perfect grounds to spread information to a broader section of society. What could circulate between a relative few can now circulate globally overnight. This advantage also marked the increase in the number of fake news attacks by its users, which is unsuitable for a healthy society. Therefore, there is a need for good algorithms to identify and take down fake information as soon as they appear. This paper aims at solving the problem by automating the process of identifying fake news using its content. Evaluation metrics like the accuracy of correct classification, precision, recall and f1-score assess the performance of the approach. The machine learning approach achieved its best performance with 96.7 percentage accuracy, 96.2 percentage precision, 97.5 percentage recall and 96.9 percentage f1 score on the ISOT dataset. 2022 IEEE. -
Farm field security system using CNN and GSM module
Loss of crops and the destruction of livestock have been a major problem for many people in rural areas due to grass-fed animals whose food is derived from plants. According to research 32% are herbivores [1]. Reduced emissions from deforestation as well as deforestation are the main reason for wildlife moving towards urban areas. It results in wildlife infestation and human and animal conflicts. Therefore, compensating for the rapid loss of crops and the slaughter of livestock requires animal shelter and isolation in order to restrict the entry of animals into farm fields. The paper describes an effective and reliable way to identify and repel wildlife from farmland and to send real-time data to the farmer indicating animal attack on fields. An image of an animal will be obtained by convolution neural networks using intensive reading algorithms that provide a message to the farmer using the GSM module. It uses a user alert system and the animal scaring method. The test results show that the proposed algorithm has high visual accuracy. The detection level of the test set is achievable and the detection result is reliable. 2024 Author(s). -
FDI in Developing Nations: Unveiling Trends, Determinants and Best Practices for India
In the recent UNCTAD World Investment Report 2023, China has the highest FDI inflows among the developing countries, following Brazil, India, Mexico, and Indonesia. These five developing countries attracted more FDI inflows in the year 2022. However, among these five countries, China and the other four countries have a lot of differences in FDI inflows. So, this study investigates the factors helping China get more FDI inflows by analyzing the trends and determinants of FDI inflows. The study also compares all the selected countries to suggest the best practices India can adopt to enhance its FDI attractiveness. So, the study considered economic indicators like GDP, infrastructure, trade openness, and natural resources. Further, panel data analysis was used to investigate the determinants influencing FDI inflows, utilizing the Panel Autoregressive Distributed Lag (P-ARDL) model for the data from 1990 to 2022. The findings showed that trade openness, market size, and quality of infrastructure explain the attraction of FDI inflows in selected countries in the long run. Thus, it is important to implement policies that encourage international collaboration by raising trade, lowering corporate expenses, and making infrastructural investments. India's availability of a large consumer market, developed infrastructure, and government initiatives like 'Make in India,' and "Skill India"have pulled major FDI inflows. India should prioritize manufacturing, IT, and healthcare while improving infrastructure and streamlining regulations. 2024 IEEE. -
Feature Based Fuzzy Framework for Sentimental Analysis of Web Data
Social mass media has emerged as a projectile platform for the evolution of web data. The sentimental Analysis where the huge textual online reviews are analyzed to extract the actual sentiment or emotions hidden in the reviews. In this paper an effective approach for sentimental analysis of web data is proposed which deploys the fuzzy based machine learning algorithm to accomplish fine-level sentiment analysis of huge online opinions by assimilating the fuzzy linguistic hedges influence on opinion descriptors. The seven layered categories are designed that uses SentiWordNet which has three stages: Pre-processing phase, Feature Selection Phase and Fuzzy based Sentiment Analysis phase. Various machine learning algorithms like AdaBoost, (IBK) K-Nearest Neighbour, (NB) Nae Bayes and (SVM)/SMO Support Vector Machine are used for classification. Jsoup is implemented for gathering web opinions which are subjected to initial processing task later applied with stemming and tagging. This fuzzy based methodology is investigated for Mobile, Laptops dataset, also compared with state-of-the-art approaches which demonstrate upper indication of 94.37% accurateness through Kappa indicators showcasing lesser error rates. The investigational outcomes are tested on training data using Ten-Fold cross validation which concludes that this approach can be efficaciously used in Sentimental analysis as an aid for online decision. 2019 IEEE. -
Feature extraction of clothing texture patterns for classification
Different features are extracted for Pattern Recognition using an efficient algorithms like Scale Invariant Feature Transform, Rotation invariant Radon Transform and extracting statistical features of a texture image. Support vector machine with RBF kernel in Weka is used in this paper for classification. This paper shows method to classify the clothing texture patterns like strips, plaid, pattern less and irregular pattern. This paper also proposes a method which can be efficient method to apply for the real time natural texture patterns and colors recognition systems. This paper gives the experiments results and the proposed method to enhance the experiments accuracy in future scope. 2015 IEEE. -
FEC & BCH: Study and implementation on VHDL
Channel encoding and Forward Error Correction is a crucial element of any communication system. This paper gives a brief overview of the fundamentals, mechanism and importance of Forward Error Correction. The design and implementation of a (63,36,5) BCH Codec is also projected in the later sections. All simulations are made on MATLAB R2018b and the VHDL implementations have been carried out using Xilinx Vivado 2018.2. 2019 IEEE -
Federated Learning and Blockchain: A Cross-Domain Convergence
Gaining significant attention within decentralized contexts, Federated Learning (FL) has been positioned as a highly desirable method for machine learning. By enabling multiple entities to train a shared model cooperatively, data privacy and security are preserved by Federated Learning. Harnessing inherent transparency and accountability of blockchain technology to trace and authenticate updates effectively in federated learning has transpired as an up-and-coming avenue to tackle data challenges related to confidentiality, protection, and reliability. This study examines the viability of federated learning and blockchain integration across multiple dimensions. The technological components of this integration., including incentive systems, consensus mechanisms, data validation, and smart contracts, are delved into. In the study, a novel proposed model for federated learning integrated with blockchain is designed and implemented. It is observed that the mean cypher size is 100 bytes for varying values of gradients. The average throughput recorded is 1.7 bytes per second, while the mean accuracy is 87.1% for 50 epochs. 2023 IEEE. -
File Validation intheData Ingestion Process Using Apache NiFi
In the industries of today, development and maintenance of data pipelines is of paramount importance. With large volumes of data being generated across industries on a continuous basis, there is a growing need to process and store this ingested data in a fast, and efficient manner. Apache NiFi is one such tool which possesses crucial capabilities that can be used to enhance, modify, and automate data pipelines. However, automation of the ingestion process creates certain inherent issues which, without being resolved, tend to be detrimental to the entire ingestion process. These issues vary in nature, ranging from corrupted data to changes in the file schema, to name a few. In this paper, a solution to this problem is proposed. By exploiting Apache NiFis custom processor development capabilities, problem-specific processors can be designed and deployed which can ensure accurate validation of the ingestion process on a real-time basis. To demonstrate this, two processors were developed as a proof-of-concept, which tackle specific file-related validation issues in the ingestion processthat of the file size, and, the ingestion frequency. These custom-built processors are designed to be inserted into the pipeline at key points to ensure that the ingested data is validated against certain standards and requirements. Having successfully demonstrated its capabilities, the paper presents the exploitation of Apache NiFis custom processor capabilities as a potential way forward to resolve the plethora of ingestion issues in industry, today. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024. -
FIN2SUM: Advancing AI-Driven Financial Text Summarization with LLMs
In the modern financial sector, the rapid digitalization of financial reports necessitates efficient and reliable text summarization tools. This research introduces FIN2SUM, a novel framework designed for summarizing the managerial analysis and discussion sections of 10-K reports from top NASDAQ-listed companies. The study aims to evaluate Large Language Models (LLMs) in financial text summarization, highlighting LLAMA-2's adeptness in processing complex financial information, thus making FIN2SUM a vital tool for analysts and decision-makers. The methodology includes a thorough evaluation of three state-of-the-art LLMs - LLAMA-2, FLAN, and Claude 2 - using BERT and ROUGE scores. The research concludes that FIN2SUM, enhanced by LLAMA-2, significantly advances AI-driven financial text summarization. 2024 IEEE. -
Financial analytical usage of cloud and appropriateness of cloud computing for certain small and medium-sized enterprises
The term "cloud computing"refers to a novel approach of providing useful ICTs to consumers over the internet on an as-needed and pay-per-usage basis. Businesses may streamline internal processes, increase contact with customers, and expand their market reach with the aid of cloud computing, which provides convenient and inexpensive access to cutting-edge information and communication technologies. Developing economies like India's present unique problems for small and medium-sized businesses (SMEs), such as a lack of funding, an inadequate workforce, and inadequate information and communication technology (ICT) use. Various advantages offered by current information and communication technology solutions are unavailable to SMEs because of these limitations. If small and medium-sized enterprises (SMEs) are seeking to enhance their internal operations, communication with customers and business partners, and market reach using current information and communication technology (ICT) solutions, cloud computing might be a good fit for them. Therefore, SMEs are particularly well-served by cloud computing. Companies with a lack of capital, personnel, or other resources to deploy and use appropriate ICTs may greatly benefit from cloud computing, and the public cloud in particular. 2024 Author(s). -
Financial Big Data Analysis Using Anti-tampering Blockchain-Based Deep Learning
This study recommends using blockchains to track and verify data in financial service chains. The financial industry may increase its core competitiveness and value by using a deep learning-based blockchain network to improve financial transaction security and capital flow stability. Future trading processes will benefit from blockchain knowledge. In this paper, we develop a blockchain model with a deep learning framework to prevent tampering with distributed databases by considering the limitations of current supply-chain finance research methodologies. The proposed model had 90.2% accuracy, 89.6% precision, 91.8% recall, 90.5% F1 Score, and 29% MAPE. Choosing distributed data properties and minimizing the process can improve accuracy. Using code merging and monitoring encryption, critical blockchain data can be obtained. 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG. -
Financial Lexicon based Sentiment Prediction for Earnings Call Transcripts for Market Intelligence
Sentiment based stock price direction detection has been an exciting study in the field of finance which is drawing a lot of attention from the investor community. Sentiments are used to detect the changes in the stock price movements for the subsequent periods. Investor community uses these sentiments derived from news, celebrity speech and events to plan trading and investment strategies. Several studies have been done in the past with sentiments, but use of Earnings Call Transcripts (ECT) has not been explored for market intelligence hitherto. Standard dictionary based lexicons like Vader, AFINN and NRC have not performed well in finance as they are domain agnostic. There is a need to develop a financial lexicon based on the ECT corpora, which may provide a better lift over the standard lexicons. This study has observed that Vader has performed poorly as opposed to the newly developed financial lexicon. Machine learning based generative lexicon engine using Bayesian approach, which is termed as FNB Lex was developed in this research study to overcome the limitations of standard domain agnostic lexicons. The lexicon development was performed on quarterly Earning Call Transcripts (ECT) of sixteen IT companies spanning over ten years. The study also investigates the detection of inverse effect in stock price movements based on the sentiments of the previous period. Machine Learning (ML) models like Naive Bayes, FNB Lex, SVM and biLSTM were developed and their discriminatory powers were assessed. NB Lex provided much better lift in detecting the inverse effect as opposed to other models. 2024 IEEE. -
Financial Market Forecasting using Macro-Economic Variables and RNN
Stock market forecasting is widely recognized as one of the most important and difficult business challenges in time series forecasting. This is mainly due to its noise. The use of RNN algorithms for funding has attracted interest from traders and scientists. The best technique for learning long-term memory sequences is to use long and short networks. Based on the literature, it is acknowledged that LSTM neural networks outperform all other models. Macroeconomics is a discipline of economics that studies the behavior of the economy as a whole. Macroeconomic factors are economic, natural, geopolitical, or other variables that influence the economy of a country. This study studies and test several macroeconomic variables and their significance on stock market forecasting. In macroeconomics, we have series that are updated once a month or even once a quarter, with data that is rarely more than a few hundred characters long. The amount of data given can sometimes be insufficient for algorithms to uncover hidden patterns and generate meaningful results. Depending on the prediction needs, we proposed a feasible LSTM design and training algorithm. According to the findings of this study, the inclusion of macroeconomic variable has a significant impact on stock price prediction. 2022 IEEE. -
Financing for SDGs in India in Post Pandemic era - Challenges & Way forward
In 2015, a resolution known as Agenda 2030 was passed by United Nations General Assembly in which seventeen goals for Sustainable Development were laid down for global dignity, peace and prosperity. The post- pandemic era became full of uncertainties in pursuing those Sustainable Development Goals (SDGs) and its implementation became a challenge especially for the developing economies like India. The country is facing a tremendous gap in arranging for resources to meet the climatic changes and attaining the SDGs. India requires 170 billion dollars per year from 2015-2030 to fulfill the Sustainable Development Goals as per the estimation done by National Determined Contribution, a body setup after Paris agreement 2015 to monitor the efforts of the country towards reducing national emissions and adapting to climate change. There is a huge concern amongst the various agencies on exploring the ways to fill this financing gap especially after the economic slowdown seen in the post pandemic era. This research paper analyses the challenges imposed by the COVID 19 pandemic on financing for SDGs and also explores the options to mitigate them. The articles and research papers related to SDG financing are reviewed by the researchers to arrive at the above mentioned statements. This paper is an attempt to draw the attention of worldwide authorities towards this grim situation as sustainable finance is far from reality in India and requires immediate up scaling. The Electrochemical Society -
Finding Real-Time Crime Detections during Video Surveillance by Live CCTV Streaming Using the Deep Learning Models
Nowadays, securing people in public places is an emerging social issue in the research of real-Time crime detection (RCD) by video surveillance, in which initial automatic recognition of suspicious objects is considered a prime problem in RCD. Dynamic live CCTV monitoring and finding real-Time crime activities by detecting suspicious objects is required to prevent unusual activities in public places. Continuous live CCTV video surveillance of objects and classification of suspicious activities are essential for real-Time crime detection. Deep training models have greatly succeeded in image and video classifications. Thus, this paper focuses on the use of trustworthy deep learning models to intelligently classify suspicious objects to detect real-Time crimes during live video surveillance by CCTV. In the experimental study, various convolutional neural network (CNN) models are trained using real-Time crime and non-crime videos. Three performance parameters, accuracy, loss, and computational time, are estimated for three variants of CNN models for the real-Time crime classifications. Three categories of videos, i.e., crime video (CV), non-crime video (NCV), and weapon-crime video (WCV), are used in the training of three deep models, CNN, 3D CNN, and Convolutional Long short-Term memory (ConvLSTM). The ConvLSTM scored higher accuracy, lower loss values, and runtime efficiency than CNN and 3D CNN when detecting real-Time crimes. 2024 ACM. -
Fine-tuning Language Models for Predicting the Impact of Events Associated to Financial News Articles
Investors and other stakeholders like consumers and employees, increasingly consider ESG factors when making decisions about investments or engaging with companies. Taking into account the importance of ESG today, FinNLP-KDF introduced the ML-ESG-3 shared task, which seeks to determine the duration of the impact of financial news articles in four languages - English, French, Korean, and Japanese. This paper describes our team, LIPIs approach towards solving the above-mentioned task. Our final systems consist of translation, paraphrasing and fine-tuning language models like BERT, Fin-BERT and RoBERTa for classification. We ranked first in the impact duration prediction subtask for French language. 2024 ELRA Language Resource Association. -
FinTech: Answer for Financial Literacy and Financial Inclusion in India
In India, financial literacy and financial inclusion is the need of the hour. Though economic growth of the country is growing in positive direction, it derailed by many factors such as financial literacy, accountability, and stability of the common public. It could be due to the deprived accessibility to the financial services in India. This study addresses the two key elements in economic growth of the country namely financial literacy and financial inclusion and how it could be handled by financial technology. This study sets up the platform in which it is trying to include perception and attitude of both the provider and the user of the fintech services and compiling its impact on both financial literacy and financial inclusion. A sample size of 644 respondents have been selected using multi-stage sampling technique and distributed with structured questionnaire. The study result gives implication for fin-tech service providers in understanding the consumer perspectives and government for policy making. The Electrochemical Society -
Fire Resistance of Concrete with Partial Replacement of Ceramic Waste and Carbon Fiber as Additives
One of the primary hazards that causes catastrophic damage to properties and peoples lives is fire. Although ceramic garbage is deposited on the land, it is a non-biodegradable waste that pollutes the environment. This study is based on the use of industrial waste products such as ceramic sanitary waste to improve the mechanical qualities of concrete that have been exposed to elevated temperatures. An experimental investigation was carried out on cubes, cylinders, and beams to assess compressive strength, split tensile strength, and flexural strength with fractional replacement of fine aggregates with 10, 20, and 30% of ceramic waste and 0, 1, and 2% of carbon fibers as additives at normal and elevated temperature as per ASTM code recommendations and the results shown as a significant improvement. The strength of M30 grade concrete with partial replacement of fine aggregate with ceramic waste up to 30% and carbon additives up to 2% shows an improvement of compressive strength by 17.56% than conventional concrete. It is also observed that normal M30 grade concrete loses its strength by 49.6% when it is exposed to 600C and with fractional replacement of fine aggregate by ceramic waste by up to 30% and carbon additives by up to 2% shows the loss of strength is decreased up to 22.67%. It shows that it is the probable substitute solution for the secure discarding of Ceramic waste. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024. -
Firefly Algorithm andDeep Neural Network Approach forIntrusion Detection
Metaheuristic optimization has grown in popularity as a way for solving complex issues that are difficult to solve using traditional methods. With fast growth of the available storage space and processing capabilities of the modern computers, the machine learning domain, that can be succinctly formulated as the process of enabling the computers to make successful forecasts based on the previous experiences, has recently been under spectacular growth. This paper presents intrusion detection approach by utilizing hybrid method between firefly algorithm and deep neural network. The basic firefly algorithm, as a frequently employed swarm intelligence method, has several known deficiencies, and to overcome them, an enhanced firefly algorithm was proposed and used in this manuscript. For experimental purposes, KDD Cup 99 and NSL-KDD datasets from Kaggle and UCL repositories were taken and comparison with other frameworks that have been validated for the same datasets was executed. Based on simulation data, proposed method was able to establish better values for accuracy, precision, recall, F-score, sensitivity and specificity metrics than other approaches. 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. -
Flight Arrival Delay Prediction Using Deep Learning
This project is aimed to solve the problem of flight delay prediction. This problem does not only affect airlines but it can cause multiple problems in different sectors i.e., commercial (Cargo aviation), passenger aviation, etc. There are a number of reasons why flights can be delayed, with weather being the main one. Our goal in this study is to forecast flight delays resulting from a variety of reasons, such as inclement weather, delayed aircraft, and other issues. The dataset gives itemized data on flight appearances and postponements for U.S. air terminals, classified via transporters. The information incorporates metrics such as the number of arriving flights, delays over 15 minutes, cancellation and diversion counts, and the breakdown of delays attributed to carriers, weather, NAS (National Airspace System), security, and late aircraft arrivals. For the purpose of predicting flight delays, the outcomes of several machine learning algorithms are examined, including Ridge, Lasso, Random Forest, Decision Tree, and Linear regression. With the lowest RMSE score of 0.0024, the Random Forest regressor performed the best across all scenarios. A deep learning model using a dense neural network is built to check how accurate a deep learning model will be while predicting the delay and the result was an RMSE score of 0.1357. 2024 IEEE.