Everyday, many individuals face online trolling and receive hate on different social media platforms like Twitter, Instagram to name a few. Often these comments involving racial abuse, hate based on religion, caste ar...
Everyday, many individuals face online trolling and receive hate on different social media platforms like Twitter, Instagram to name a few. Often these comments involving racial abuse, hate based on religion, caste are made by anonymous people over the internet, and it is quite a task to keep these comments under control. So, the objective was to develop a Machine Learning Model to help identify these comments. A Deep Learning Model (a sequential model) was made and it was trained to identify and classify a comment based on whether it is an apt comment or not. LSTM (Long Short-Term Memory) is a type of recurrent neural network (RNN) that is particularly well-suited for modeling sequential data, such as text. LSTMs are capable of modeling long-term dependencies in sequential data. In the case of text classification, this means that LSTMs can take into account the context of a word or phrase within a sentence, paragraph, or even an entire document. LSTMs can learn to selectively forget or remember information from the past, which is useful for filtering out noise or irrelevant information in text. LSTMs are well-established in the field of natural language processing (NLP) and have been shown to be effective for various NLP tasks, including sentiment analysis and text classification. Binary cross-entropy is a commonly used loss function in deep learning models for binary classification problems, such as predicting whether a comment is toxic or not. Binary cross-entropy is designed to optimize the model's predictions based on the binary nature of the classification task. It penalizes the model for assigning a low probability to the correct class and rewards it for assigning a high probability to the correct class. The loss function is differentiable, which allows gradient-based optimization methods to be used during training to minimize the loss and improve the model's performance. Binary cross-entropy is a well-established loss function that has been extensively used
A phone number is a personal unique number which can identify its owner. Personal information can be exploited through a person’s phone number. Many methods such as phishing and spam attacks have been used to exploit...
A phone number is a personal unique number which can identify its owner. Personal information can be exploited through a person’s phone number. Many methods such as phishing and spam attacks have been used to exploit a person’s personal information. Nowadays, a phone number has many linked important information such as bank accounts, and passwords which make it more vulnerable to attacks. The growing demand for accessibility and connectivity is creating demand for an efficient solution. Hence, a caller identification app which can identify caller ID and spam protection of the incoming call or SMS is required. It protects users from attending unwanted calls and messages. In this work, the design and working of a caller identification app using AWS services is explored. Many researchers have conducted studies and surveys to identify the process of caller identification and develop caller identification methods and have published their research about the same. Most of these methods are conventional. What makes the proposed model novel, is that it uses a cloud-based new age model with complete backend design to develop a strong caller identification application with its roots in the AWS cloud. Though other models have been successful, this model is better because of the benefits provided by the cloud, which include high availability, scalability, elasticity, security, and cost-friendliness.
This study presents a novel machine learning approach, Adaptive Green Space Quality Transfer Learning (AGSQTL), for assessing urban green space quality. Unlike traditional transfer learning methods, AGSQTL incorporate...
详细信息
The world's largest producers of sugarcane, which is used to make both sugar and bioethanol, are Brazil and India. The crop is primarily grown in tropical and subtropical regions. These nations produce 40% of the ...
详细信息
ISBN:
(数字)9798331518592
ISBN:
(纸本)9798331518608
The world's largest producers of sugarcane, which is used to make both sugar and bioethanol, are Brazil and India. The crop is primarily grown in tropical and subtropical regions. These nations produce 40% of the world's bioethanol and 80% of its sugar. Sugarcane is susceptible to a wide range of diseases, including bacterial blight, mosaic, red rot, rust, and yellow, due to a multitude of environmental factors. An accurate and dependable automated disease detection system is needed for sugarcane disease diagnosis. In this work, we created the convolution neural network model EfficientNetB0 utilizing data augmentation, evaluated its effectiveness in identifying diseases affecting sugarcane plants using leaf pictures, and compared its performance. The precision, recall, and F1 score values of the EfficientNetB0 model are 0.9812, 0.9821, and 0.9816. This model is a helpful tool that growers may use to prevent sugarcane infections from interfering with the harvesting process, as it has an accuracy rate of 98.12%.
With the increasing availability of data, machine learning (ML) predictive models have become a popular tool for making informed decisions in various fields. However, choosing suitable algorithms and techniques to dev...
With the increasing availability of data, machine learning (ML) predictive models have become a popular tool for making informed decisions in various fields. However, choosing suitable algorithms and techniques to develop accurate forecasting models is a challenging task. In this study, we aim to compare different ML algorithms and methods to develop accurate prediction models. We analyzed the results to determine the most effective approach to developing accurate forecasting models. In addition, we investigated the effect of hyperparameter tuning on the performance of this model. The results of this study can provide valuable insights for practitioners and researchers in choosing the most appropriate machine learning algorithms and methods to develop accurate prediction models. Our study shows the impact of choosing the right features and choosing the method of data pre-processing and hyperparameter tuning on model performance. Finally, our results can inform the design of more accurate prediction models and contribute to the development of the field of machine learning.
Internet of Things (IoT) applications, such as e-healthcare departments have grown tremendously where devices gather patient data and instantly transmit it over a distance to servers. Despite its huge advantages, IoT ...
Internet of Things (IoT) applications, such as e-healthcare departments have grown tremendously where devices gather patient data and instantly transmit it over a distance to servers. Despite its huge advantages, IoT in the healthcare sector has acquired little consideration, largely because of the threats of unauthorized access to private health data made possible by the weak wireless communication channel. The economic complexity of the current security protocols makes them inappropriate, hence new security protocols must be developed for resource-constrained and diverse IoT networks. This work suggests an authenticated key agreement protocol for the Internet of medical things (IoMT) that is based on zero-knowledge proofs (ZKP). The research will include the comparison between computational delays in traditional protocols and ZKP protocol. This paper provides a protocol to obtain privacy of data in IoMT environment on the basis of zero knowledge phenomenon and anonymous communication by ZKP.
This research paper presents an innovative approach for automated bone fracture discovery and bracket using digital image processing ways and the Scale Invariant point Transform (SIFT) algorithm. The proposed system e...
This research paper presents an innovative approach for automated bone fracture discovery and bracket using digital image processing ways and the Scale Invariant point Transform (SIFT) algorithm. The proposed system excerpts distinctive original features from radiographic images, enabling robust identification of fracture patterns. Through a multi-step process, including pre-processing, point birth, and bracket, the system achieves high delicacy in differencing between colourful fracture types. Experimental results demonstrate the effectiveness of the system on a different dataset, showcasing its eventuality for enhancing clinical diagnostics and expediting treatment planning. This exploration contributes to the advancement of computer- backed medical imaging systems, offering a promising tool for accurate and effective fracture opinion.
In today's digital world, the internet deeply ingrains in all facets of our lives. People highly depend on myriad of online sources for news. The proliferation of fake news is increasingly recognized as a signific...
详细信息
Cybersecurity has seen widespread adoption across various domains, encompassing critical business infrastructure, residential settings, personal devices, and machinery. This has led to the development of innovative ca...
Cybersecurity has seen widespread adoption across various domains, encompassing critical business infrastructure, residential settings, personal devices, and machinery. This has led to the development of innovative capabilities to effectively address security challenges within the realm of Internet of Things (IoT) devices. Notably, new security measures, including the application of deep learning for intrusion detection, have been introduced. Recent research primarily focuses on enhancing algorithms for a deeper understanding of IoT security. This research explores intrusion detection techniques employing deep knowledge mining. It involves a comprehensive analysis of various deep knowledge mining methods, comparing their overall performance. The objective is to identify a precise approach for implementing intrusion detection in the IoT context. The research leverages deep learning methodologies, specifically convolutional neural networks (CNN), long short-term memory (LSTM), and gated recurrent units (GRU). Furthermore, it incorporates an extensive dataset tailored for IoT intrusion detection. The empirical results of this study are thoroughly examined and compared against existing IoT intrusion detection techniques. Notably, the proposed approach demonstrates superior accuracy when compared to other methods.
Chatbots are applications which simulate human conversational patterns, engage in human-to-human communication, and respond to user queries using data from their training set. There are two categories of chatbots: rul...
Chatbots are applications which simulate human conversational patterns, engage in human-to-human communication, and respond to user queries using data from their training set. There are two categories of chatbots: rule-based and self-learning. One of the beneficial things a company or organisation may have is a chatbot. They can be utilised for customer service and query resolution in many different industries. In this paper, research is done on NLP and a model is designed to build a chatbot using NLP techniques. The chatbot built using this model can be used as a normal conversation bot or as a customer support tool.
暂无评论