2024 – 2025 IEEE python Machine Learning Projects

The increasing frequency of network-based attacks and security breaches underscores the critical need for robust Intrusion Detection Systems (IDS) to safeguard computer networks and sensitive information. This project aims to develop an advanced IDS integrated with a Decoy system, utilizing the powerful Random Forest and K-Nearest Neighbors (KNN) machine learning algorithms. The primary goal is to design and implement an intelligent IDS capable of effectively detecting and classifying various types of network intrusions while minimizing false positives.To bolster the system’s security, an integrated Decoy system will be employed to deceive potential attackers and divert their attention from actual network resources. The IDS will leverage the Random Forest algorithm, an ensemble learning method that combines multiple decision trees to achieve high accuracy and robustness. Additionally, the project will investigate the use of the K-Nearest Neighbors (KNN) algorithm, a non-parametric technique that classifies new data points based on the majority vote of their nearest neighbors, for intrusion detection.

As the world responded to the Coronavirus Disease 2019 (COVID-19) pandemic in 2020, digital operations became more important, and people started to depend on new initiatives such as the cloud and mobile infrastructure. Consequently, the number of cyberattacks such as phishing has increased. Phishing websites can be detected using machine learning by classifying the websites into legitimate or illegitimate websites. The purpose of the study is to conduct a mini-review of the existing techniques and implement experiments to detect whether a website is malicious or not. The dataset consists of 11,055 observations and 32 variables. Three supervised learning models are implemented in this study: Decision Tree, K-Nearest Neighbour (KNN), and Random Forest. The three algorithms are chosen because it provides a better understanding and more suitable for the dataset. Based on the experiments undertaken, the result shows Decision Tree has an accuracy of 91.16% which is the lowest compared to the other models, 97.6% for the KNN model which is the highest among all the models and 94.44% accuracy for the Random Forest model. Through comparisons between the three models, KNN was the prime candidate for the best model considering that it has the highest accuracy. However, Random Forest is deemed more suitable for the dataset even though the accuracy is lesser because of the lowest false-negative value than the other models. The experiments can be further investigated with different datasets and models for comparative analysis.

Fiber optic networks (FONs) are considered the backbone of telecom companies worldwide. However, the network elements of FONs are scattered over a wide area and managed through a centralized controller based on intelligent devices and the internet of things (IoT), with actuators used to perform specific tasks at remote locations. During the COVID-19 pandemic, many telecom companies advised their employees to manage the network using the public internet (e.g., working from home while connected to an IoT network). Theses IoT devices mostly have weak security algorithms that are easily taken-over by hackers, and therefore can generate Distributed Denial of Service (DDoS) attacks in FONs. A DDoS attack is one of the most severe cyberattack types, and can negatively affect the stability and quality of managing networks. Nowadays, software-defined networks (SDN) constitute a new approach that simplifies how the network can be managed through a centralized controller. Moreover, machine learning algorithms allow the detection of incoming malicious traffic with high accuracy. Therefore, combining SDN and ML approaches can lead to detecting and stopping DDoS attacks quickly and efficiently, especially compared to traditional methods. In this paper, we evaluated six ML models: Logistic Regression, K-Nearest Neighbor, Support Vector Machine, Naive Bayes, Decision Tree, and Random Forest. The accuracy reached 100% while detecting DDoS attacks in FON with two approaches: (1) using SVM with three features (SOS, SSIP, and RPF) and (2) using Random Forest with five features (SOS, SSIP, RPF, SDFP, and SDFB). The training time for the first approach was 14.3 s, whereas the second approach only requires 0.18 s; hence, the second approach was utilized for deployment.

Islamophobic hate speech is the indiscriminate negative attitude and behavior towards Muslims and Islam. Speech indicating prejudice against Muslims has negatively impacted the perceptions of Islam. Online platforms like Twitter have carved out policies to stop users from promoting Islamophobic hate speech, however, such content still exists which causes problems for Muslim communities globally. Hence, it becomes pivotal to find solutions to eradicate such speech from social media platforms. This paper presents an effective methodology for Islamophobic hate speech identification in online tweets using deep learning techniques. The proposed technique relies on feature extraction using a one-dimensional Convolutional Neural Network and classification using Long Short-Term Memory network based classifier. The proposed technique is validated on a dataset comprising of 1290 preprocessed online tweets and an accuracy of more than 90% is reported.

Denial of service (DoS) attack is a typical and extremely destructive attack, which poses a serious threat to the Internet security and is highly concealed, making it difficult to detect. In response to this problem, the paper proposes an efficient DoS attack traffic detection method, Random Forest and Multilayer Perceptron hybrid network attack detection algorithm (RF-MLP). At first, it is adopted that the random forest algorithm can be used for feature selection and the optimal threshold can be determined by drawing a learning curve; therefore the optimal feature subset is determined. &en the optimal feature subset is used as the input of the multilayer perceptron for training. We will analyze the experimental results obtained using different configurations by varying the number of training neurons and the number of hidden layers of the multilayer perceptron network in order to improve the accuracy and reduce the number of false results. Using the real network traffic CICIDS2017 dataset and UNSW-NB15 dataset to evaluate the method in this paper, the results show that the model can effectively detect and classify DoS attacks, the accuracy rate can reach 99.83% and 93.51%, and there is also a significant reduction in the false alarm rate, verifying the effectiveness of the method and its ease of use.

Cardiotocogram (CTG) is the most widely used in the clinical routine evaluation of the main approach to detect fetal state. In this paper, twelve machine learning single models have firstly experimented on CTG dataset. Secondly, the soft voting integration method is used to integrate the four best models to build the Blender Model, and compared with the stacking integration method. Compared with the traditional machine learning models, the model proposed in this paper performed excellently in various Classification Model evaluations, with an accuracy rate of 0.959, an AUC of 0.988, a recall rate of 0.916, a precision rate of 0.959, a F1 of 0.958 and a MCC of 0.886.

Since coronavirus has shown up, inaccessibility of legitimate clinical resources is at its peak, like the shortage of specialists and healthcare workers, lack of proper equipment and medicines etc. The entire medical fraternity is in distress, which results in numerous individual’s demise. Due to unavailability, individuals started taking medication independently without appropriate consultation, making the health condition worse than usual. As of late, machine learning has been valuable in numerous applications, and there is an increase in innovative work for automation. This paper intends to present a drug recommender system that can drastically reduce specialists heap. In this research, we build a medicine recommendation system that uses patient reviews to predict the sentiment using various vectorization processes like Bow, TF-IDF, Word2Vec, and Manual Feature Analysis, which can help recommend the top drug for a given disease by different classification algorithms. The predicted sentiments were evaluated by precision, recall, f1score, accuracy, and AUC score. The results show that classifier LinearSVC using TF-IDF vectorization outperforms all other models with 93% accuracy.

Fake review detection and its elimination from the given dataset using different Natural Language Processing (NLP) techniques is important in several aspects. In this article, the fake review dataset is trained by applying two different Machine Learning (ML) models to predict the accuracy of how genuine are the reviews in a given dataset. The rate of fake reviews in E-commerce industry and even other platforms is increasing when depend on product reviews for the item found online on different websites and applications. The products of the company were trusted before making a purchase. So this fake review problem must be addressed so that these large E-commerce industries such as Flipkart, Amazon, etc. can rectify this issue so that the fake reviewers and spammers are eliminated to prevent users from losing trust on online shopping platforms. This model can be used by websites and applications with few thousands of users where it can predict the authenticity of the review based on which the website owners can take necessary action towards them. This model is developed using Naïve Bayes and random forest methods. By applying these models one can know the number of spam reviews on a website or application instantly. To counter such spammers, a sophisticated model is required in which a need to be trained on millions of reviews. In this work “amazon Yelp dataset” is used to train the models and its very small dataset is used for training on a very small scale and can be scaled to get high accuracy and flexibility.

In recent years, the need for security of personal data is becoming progressively important. In this regard, the identification system based on fusion of multibiometric is most recommended for significantly improving and achieving the high performance accuracy. The main purpose of this paper is to propose a hybrid system of combining the effect of tree efficient models: Convolutional neural network (CNN), Softmax and Random forest (RF) classifier based on multi-biometric fingerprint, finger-vein and face identification system. In conventional fingerprint system, image pre-processed is applied to separate the foreground and background region based on K-means and DBSCAN algorithm. Furthermore, the features are extracted using CNNs and dropout approach, after that, the Softmax performs as a recognizer. In conventional fingervein system, the region of interest image contrast enhancement using exposure fusion framework is input into the CNNs model. Moreover, the RF classifier is proposed for classification. In conventional face system, the CNNs architecture and Softmax are required to generate face feature vectors and classify personal recognition. The score provided by these systems is combined for improving Human identification. The proposed algorithm is evaluated on publicly available SDUMLA-HMT real multimodal biometric database using a GPU based implementation. An experimental result on the datasets has shown significant capability for identification biometric system. The proposed work can offer an accurate and efficient matching compared with other system based on unimodal, bimodal, multimodal characteristics.

Abstract

In present-day Autism Spectrum Disorder (ASD) is gaining its momentum faster than ever. Detecting autism traits through screening tests is very expensive and time-consuming. With the advancement of artificial intelligence and machine learning (ML), autism can be predicted at a quite early stage. Though a number of studies have been carried out using different techniques, these studies didn’t provide any definitive conclusion about predicting autism traits in terms of different age groups. The proposed model was evaluated with AQ- 10 dataset and 1000 real datasets collected from people with and without autistic traits. The evaluation results showed that the proposed prediction model provides better results in terms of accuracy, specificity, sensitivity, precision, and false-positive rate (FPR) for both kinds of datasets.

Abstract:

Machine Learning algorithms sprawl their application in various fields relentlessly. Software Engineering is not exempted from that. Software bug prediction at the initial stages of software development improves the important aspects such as software quality, reliability, and efficiency and minimizes the development cost. In majority of software projects which are becoming increasingly large and complex programs, bugs are serious challenge for system consistency and efficiency. In our approach, three supervised machine learning algorithms are considered to build the model and predict the occurrence of the software bugs based on historical data by deploying the classifiers Logistic regression, Naïve Bayes, and Decision Tree. Historical data has been used to predict the future software faults by deploying the classifier algorithms and make the models a better choice for predictions using random forest ensemble classifiers and validating the models with K-Fold cross validation technique which results in the model effectively working for all the scenarios.

Abstract:

In this project, two predictive models have been designed namely students’ assessments grades and final students’ performance. The models can be used to detect the factors that influence students’ earning achievement in Universities. With rapid advancements in technology, artificial intelligence has recently become an effective approach in the evaluation and testing of student performance in online courses. Many researchers applied machine learning to predict student performance

Abstract:

Thyroid disease is one of the most common diseases among the female mass in Bangladesh. Hypothyroid is a common variation of thyroid disease. It is clearly visible that hypothyroid disease is mostly seen in female patients. Most people are not aware of that disease as a result of which, it is rapidly turning into a critical disease. It is very much important to detect it in the primary stage so that doctors can provide better medication to keep itself turning into a serious matter. Predicting disease in machine learning is a difficult task. Machine learning plays an important role in predicting diseases. Again distinct feature selection techniques have facilitated this process prediction and assumption of diseases. There are two types of thyroid diseases namely 1. Hyperthyroid and 2.Hypothyroid. Here, in this paper, we have attempted to predict hypothyroid in the primary stage. To do so, we have mainly used three feature selection techniques along with diverse classification techniques. Feature selection techniques used by us are Recursive Feature Selection(RFE), Univariate Feature Selection(UFS) and Principal Component Analysis(PCA) along with classification algorithms named Support Vector Machine(SVM), Decision Tree(DT), Random Forest(RF), Logistic Regression(LR) and Naive Bayes(NB).   Thus it’s deduced from our research that RFE helps each classifier to attain better accuracy than all the other feature selection methods used.

Abstract:

Agriculture is an important application in India. Modern technologies can change the situation of farmers and decision-making in an agricultural field in a better way. Python is used as a front end for analyzing the agricultural data set. Jupyter Notebook is the data mining tool used to predict crop production. The parameter includes in the data set are precipitation, temperature, reference crop, evapotranspiration, area, production, and yield for the season from January to December for the years 2000 to 2018. The data mining techniques like K-Means Clustering, KNN, SVM, and Bayesian network algorithm where high accuracy can be achieved.

Abstract

A non-communicable disease Diabetes is increasing day by day at an alarming rate all over the world and it may cause some long-term issues such as affecting the eyes, heart, kidneys, brain, feet and nerves. It is really important to find an effective way of predicting diabetes before it turns into one of the major problems for the human being. If we take proper precautions on the early stage, it is possible to take control of diabetes disease. In this analysis, 340 instances have been collected with 26 features of patients who have already been affected by diabetes with various symptoms categorized by two types namely Typical symptoms and Non-typical symptoms.

 Abstract:

The main objective of this paper is to find the best model to predict the value of the stock market. During the process  Of considering various techniques and variables that must be taken into account, we found out that techniques like random forest, support vector machine were not exploited fully. In, this paper we are going to present and review a more feasible method to predict the stock movement with higher accuracy. The first thing we have taken into account is the dataset of the stock market prices from previous year. The dataset was pre-processed and tuned up for real analysis. Hence, our paper will also focus on data preprocessing of the raw dataset. Secondly, after pre-processing the data, we will review the use of random forest, support vector machine on the dataset and the outcomes it generates. In addition, the proposed paper examines the use of the prediction system in real-world settings and issues associated with the accuracy of the overall values given. The paper also presents a machine-learning model to predict the longevity of stock in a competitive market. The successful prediction of the stock will be a great asset for the stock market institutions and will provide real-life solutions to the problems that stock investors face.

Abstract:

Health care field has a vast amount of data, for processing those data certain techniques are used. Data mining is one of the techniques often used. Heart disease is the Leading cause of death worldwide. This System predicts the arising possibilities of Heart Disease. The outcomes of this system provide the chances of occurring heart disease in terms of percentage. The datasets used are classified in terms of medical parameters. This system evaluates those parameters using data mining classification technique. The datasets are processed in python programming using two main Machine Learning Algorithm namely Decision Tree Algorithm and Naive Bayes Algorithm which shows the best algorithm among these two in terms of accuracy level of heart disease.

Abstract:

Since corona virus has shown up, inaccessibility of legitimate clinical resources is at its peak, like the shortage of specialists and healthcare workers, lack of proper equipment and medicines etc. The entire medical fraternity is in distress, which results in numerous individual’s demise. Due to unavailability, individuals started taking medication independently without appropriate consultation, making the health condition worse than usual. As of late, machine learning has been valuable in numerous applications, and there is an increase in innovative work for automation. This paper intends to present a drug recommender system that can drastically reduce specialists heap. In this research, we build a medicine recommendation system that uses patient reviews to predict the sentiment using various vectorization processes like Bow, TF-IDF,Word2Vec, and Manual Feature Analysis, which can help recommend the top drug for a given disease by different classification algorithms. The predicted sentiments were evaluated by precision, recall, f1score, accuracy, and AUC score.

Abstract:

Stress disorders are a common issue among working IT professionals in the industry today. With changing lifestyles and work cultures, there is an increase in the risk of stress among the employees. Though many industries and corporates provide mental health-related schemes and try to ease the workplace atmosphere, the issue is far from control. In this paper, we would like to apply machine learning techniques to analyze stress patterns in working adults and to narrow down the factors that strongly determine stress levels. Towards this, data from the OSMI mental health survey 2017 responses of working professionals within the tech industry was considered. Various Machine Learning techniques were applied to train our model after due data cleaning and preprocessing. The accuracy of the above models was obtained and studied comparatively. Boosting had the highest accuracy among the models implemented. By using Decision Trees, prominent features that influence stress were identified as gender, family history, and availability of health benefits in the workplace. With these results, industries can now narrow down their approach to reduce stress and create a much comfortable workplace for their employees.

Abstract:

Nowadays, a massive amount of reviews is available online. Besides offering a valuable source of information, these informational contents generated by users, also called User Generated Contents (UGC) strongly impact the purchase decision of customers. As a matter of fact, a recent survey revealed that 67.7% of consumers are effectively influenced by online reviews when making their purchase decisions. The consumers want to find useful information as quickly as possible. However, searching and comparing text reviews can be frustrating for users as they feel submerged with information. Indeed, the massive amount of text reviews as well as its unstructured text format prevent the user from choosing a product with ease. The star-rating, i.e. stars from 1 to 5 on Amazon, rather than its text content gives a quick overview of the product quality. This numerical information is the number one factor used in an early phase by consumers to compare products before making their purchase decision. However, many product reviews are not accompanied by a correct scale rating system, some of reviewers given good review but the rating is low and bad review but high ratings so we might get some wrong product to overcome this problem we proposed model using Natural language processing (NLP), and Deep Learning Algorithm. Natural language processing (NLP) is a subfield of linguistics, computer science, and artificial intelligence concerned with the interactions between computers and human language, in particular how to program computers to process and analyze large amounts of natural language data. NLP is used for Text processing here, then we used Recurrent Neural Network (RNN) based Long short-term memory (LSTM) for classify the Ratings based on Reviews. Long short-term memory (LSTM) is an artificial recurrent neural network (RNN) architecture used in the field of deep learning. LSTM networks are well-suited to classifying, processing and making predictions based on text data.

Abstract:

With the increasing demand of social life in today’s world one famous platform known to be Twitter plays an important role for every citizen to connect socially, whether in the form of tweeting a tweet for another person or exploring various fields in the running world. But this platform, now a day gets infected by some spammers, with the intention of increasing traffic in their spam websites they connect their URL with the informative tweet where there is no relation between the content present in the URL and the tweet message, these are called spammed tweets. This paper provides a unique approach to detect whether the tweet mentioned by the user is spam or not spam using the encoder-decoder technique used with vectorizer converter on the mentioned tweets and on its linked URL, which leads to the prediction of similarity score between them.

Abstract:

Today’s world is solely dependent on social media. A number of mental disorders caused by the social networks are seen such as cyber Addiction, net Compulsions,s, etc. this becomes more problematic when there is delayed clinical intervention. We are preparing a project where it can detect in the early stages.  We take face images of a particular user where each image depicts an expression and is categorized as positive or negative. We have 6 expressions. They are sad, normal, happy, angry, surprise, disgust. Where expressions such as sad, angry, and disgust are concluded as negative and normal, happiness and surprise are concluded as positive. For certain people, if we are having a more negative depiction then it is concluded to be a mental disorder. We use image processing with Convolution Neural Networks for detection and prediction.

Abstract:

Women and girls have been experiencing a lot of violence and harassment in public places in various cities starting from stalking and leading to sexual harassment or sexual assault. This research paper basically focuses on the role of social media in promoting the safety of women in Indian cities with special reference to the role of social media websites and applications including Twitter platforms Facebook and Instagram. This paper also focuses on how a sense of responsibility on part of Indian society can be developed for the common Indian people so that we should focus on the safety of women surrounding them. Tweets on Twitter which usually contain images and text and also written messages and quotes which focus on the safety of women in Indian cities can be used to read a message amongst the Indian Youth Culture and educate people to take strict action and punish those who harass the women. Twitter and other Twitter handles which include hashtag messages that are widely spread across the whole globe sir as a platform for women to express their views about how they feel while we go out for work or travel in public transport and what is the state of their mind when they are surrounded by unknown men and whether these women feel safe or not?

Abstract:

With the rapid development of Internet technology and social networks, a large number of comment texts are generated on the Web. In the era of big data, mining the emotional tendency of comments through artificial intelligence technology is helpful for the timely understanding of network public opinion. The technology of sentiment analysis is a part of artificial intelligence, and its research is very meaningful for obtaining the sentiment trend of the comments. The essence of sentiment analysis is the text classification task, and different words have different contributions to classification. In the current sentiment analysis studies, distributed word representation is mostly used. However, distributed word representation only considers the semantic information of the word, but ignores the sentiment information of the word. In this paper, an improved word representation method is proposed, which integrates the contribution of sentiment information into the traditional TF-IDF algorithm and generates weighted word vectors. The weighted word vectors are input into bidirectional long short term memory (BiLSTM) to capture the context information effectively, and the comment vectors are better represented. The sentiment tendency of the comment is obtained by a feed-forward neural network classifier. Under the same conditions, the proposed sentiment analysis method is compared with the sentiment analysis methods of RNN, CNN, LSTM, and NB. The experimental results show that the proposed sentiment analysis method has higher precision, recall, and F1 score. The method is proved to be effective with high accuracy on comments.

Abstract:

we are living in a post modern era and there are tremendous changes happening to our daily routines which make an impact on our health positively and negatively. As a result of these changes various kind of diseases are enormously increased. Especially, heart disease has become more common these days. The life of people is at a risk. Variation in Blood pressure, sugar, pulse rate etc. can lead to cardiovascular diseases that include narrowed or blocked blood vessels. It may causes Heart failure, Aneurysm, Peripheral artery disease, Heart attack, Stroke and even sudden cardiac arrest. Many forms of heart disease can be detected or diagnosed with different medical tests by considering family medical history and other factors. But, the prediction of heart diseases without doing any medical tests is quite difficult. The aim of this project is to diagnose different heart diseases and to make all possible precautions to prevent at early stage itself with affordable rate. We follow ‘Data mining’ technique in which attributes are fed in to SVM, Random forest, KNN, and ANN classification Algorithms for the prediction of heart diseases. The preliminary readings and studies obtained from this technique is used to know the possibility of detecting heart diseases at early stage and can be completely cured by proper diagnosis.

Abstract :

Human Activity Recognition (HAR) based on sensor networks is an important research direction in the fields of pervasive computing and body area network. Existing researches often use statistical machine learning methods to manually extract and construct features of different motions. However, in the face of extremely fast-growing waveform data with no obvious laws, the traditional feature engineering methods are becoming more and more incapable. With the development of Deep Learning technology, we do not need to manually extract features and can improve the performance in complex human activity recognition problems. By migrating deep neural network experience in image recognition, we propose a deep learning model (InnoHAR) based on the combination of Inception Neural Network and recurrent neural network. The model inputs the waveform data of multi-channel sensors end-to-end. Multi-dimensional features are extracted by Inception-like modules with using of various kernel-based convolution layers. Combined with GRU, modeling for time series features is realized, making full use of data characteristics to complete classification tasks. Through experimental verification on three most widely used public HAR datasets, our proposed method shows consistent superior performance and has good generalization performance, when compared with state-of-the-arts.