.png)
The A-Z Guide to Research Work is designed to provide researchers, academics, and enthusiasts with a roadmap for conducting research and implementing findings in Python. Whether you're a novice researcher looking to learn the ropes or an experienced practitioner seeking to expand your skill set, this repository offers a wealth of resources to support your research endeavors.
1-Comprehensive Coverage: Explore a wide range of research topics, methodologies, and techniques, covering various fields such as computer science, engineering, mathematics, and more.
2-Implementation of Research Papers: Dive into practical implementations of research papers using Python, including algorithms, models, simulations, and experiments, alongside detailed explanations and code examples.
3-Hands-On Projects: Engage in hands-on projects that demonstrate the application of research methodologies in real-world scenarios, fostering a deeper understanding of research principles and methodologies.
4-Supplementary Resources: Access supplementary materials, including articles, tutorials, datasets, and curated resources, to enrich your learning experience and stay updated with the latest developments in research.
Research Methodologies: Covering fundamental research methodologies such as literature review, experimental design, data collection, analysis, and interpretation.
Implementation of Research Papers: Demonstrating practical implementations of research papers in various domains, including machine learning, data analysis, optimization, and more.
Explore the repository's contents, delve into research methodologies and implementations, and leverage the provided code examples, projects, and resources to enhance your research skills and proficiency.
We believe that the most effective learning and growth happen when people come together to exchange knowledge and ideas. Whether you're an experienced professional or just beginning your machine learning Research journey, your input can be valuable to the community. We welcome contributions from the community! Whether it's fixing a bug, adding a new Research ideas implementation, or improving documentation, your contributions are valuable. Please contact on my skype ID: themushtaq48 for guidelines on how to contribute.
Why Contribute?
1- Share Your Expertise: If you have knowledge or insights in machine learning or Research, your contributions can assist others in learning and applying these concepts.
2- Enhance Your Skills: Contributing to this project offers a great opportunity to deepen your understanding of machine learning Research paper. Writing, coding, or reviewing content will reinforce your knowledge while uncovering new areas of the field.
3- Collaborate and Connect: Join a community of like-minded individuals committed to advancing AI education. Work with peers, receive feedback, and build connections that may open up new opportunities.
4- Make a Difference: Your contributions can shape how others learn and engage with machine learning. By refining and expanding content, you help shape the education of future engineers and AI experts.
If you want to contact me, you can reach me through social handles.
.png){kind=link}
Contributions are welcome! Whether it's implementing new research papers, improving existing content, or providing feedback, your contributions can help enrich the learning experience for others. Please contact to my skype ID:themushtaq48 for guidelines on how to contribute.
Together, let's make this the best Research learning hub repository! 🚀
Star this repo if you find it useful ⭐
https://coursesteach.com/
- Essential Apps and Websites for Every PhD Student
- How to respond to comments received in Peer Review
- A Comprehensive Guide on How to Write a Research Proposal
- Finding the Right Journal for Your Research Paper: A Comprehensive Guide
- How To Read Research Papers
- Mastering the Art of Writing a Research Paper: A Comprehensive Guide
- Virtual University Thesis Tempalate
- Mastering APA Style: A Guide to Research Paper Writing
| Journal Name | Subject\Area | ISSN # | Country | Frequency | Impact Factor | Category | Publisher | Open Access | Accptance/Rejection Rate | Paper |
|---|---|---|---|---|---|---|---|---|---|---|
| 🌐1- Expert Systems with Applications | Computer Science, Engineering | 0957-4174 | United Kingdom | 8 Months | 8.5 | SCIE,W | Enhanced deep learning algorithm development to detect pain intensity from facial expression images | |||
| 🌐2- IEEE Transactions on Cybernetics | Computer Science, Engineering | 2168-2275,2168-2267 | United States | NA | 11.8 | SCIE,W | Deep Pain: Exploiting Long Short-Term Memory Networks for Facial Expression Classification |
| Title | Published Date | Research Questions | Model performance metrics | Research Gap | Taking Notes | Input/Target Features | Journal name/ Category | Limitations | Future Directions |
|---|---|---|---|---|---|---|---|---|---|
| 🌐1- Identifying Interrelated Factors of Fatal and Injury Traffic Accidents Using Association Rules | 2023 | What are the Interrelated Factors of Fatal Injury | Apriori algorithm (Support=0.05, Confidence=0.70 and Lift >1) | NA | Introduction, Methodology and the description of confidence, support and lift | ||||
| 🌐2- Road traffic accidents analysis using association rule mining and descriptive analytics | 2023 | What are the characteristics of road traffic accidents | FP Growth Algorithm (Support=0.45, Confidence=0.95 and Lift >1) | NA | Introduction | ||||
| 🌐3- A Novel Approach to Avoid Road Traffic Accidents and Develop Safety Rules for Traffic Using Crash Prediction Model Technique | 2023 | How can Crash Prediction Models (CPMs) developed through machine learning approaches contribute to minimizing road traffic accidents and developing effective safety rules for traffic? | Random Forest achieves the highest values of accuracy and precision of around 60% | NA | Introduction, Related work | ||||
| 🌐4- A NOVEL ROAD TRAFFIC ACCIDENTS PREDICTION MODEL WITH RANDOM CLASSIFIER AFTER HYPER-PARAMETER TUNED USING GRIDSEARCHCV | 2023 | How can we effectively predict road traffic accidents using a novel prediction model incorporating Random Forest Classifier after hyper-parameter tuning with GridsearchCV? | Gradient Boosting Classifier (84.9 Accuracy) | NA | Future work | ||||
| 🌐5- Fatality Prediction for Motor Vehicle Collisions: Mining Big Data Using Deep Learning and Ensemble Methods | 2023 | How effective are deep learning and ensemble methods in predicting the fatality outcome of motor vehicle collisions using large-scale datasets? | Neural Network (75% Accuracy) | NA | Introduction |
| 📰 Final Synopsis | Defance Date | Research Questions | Research Gap | Dataset | 📔Notebook | Blog | Diagram | Final Thesis |
|---|---|---|---|---|---|---|---|---|
| Synopis file | June-14-2024 | 1. Can ML algorithms effectively model the injury severity level resulting from collisions in road accidents, and if so, which ML classifier offers optimal performance in predicting the injury severity level across different geographic regions? 2- What are the underlying hidden factors that contribute to fatal or major injuries? 3- Given the differences in road safety regulations and enforcement between Pakistan and Canada, how do the identified risk factors and their impacts on injury severity differ between these two countries? What policy recommendations can be derived from these differences to enhance road safety initiatives in both regions? | Research Gap | 1-2 |  |
Medium | 1-2 | Final Thesis Files |
| f1 | precision | recall | accuracy | training_time | inference_time | |
|---|---|---|---|---|---|---|
| NB | 0.830188679245283 | 0.8461538461538461 | 0.8148148148148148 | 0.8085106382978723 | 0.0032272338867187 | 0.0007801055908203 |
| LR | 0.8518518518518519 | 0.8846153846153846 | 0.8214285714285714 | 0.8297872340425532 | 0.0356471538543701 | 0.0002150535583496 |
| KNN | 0.7058823529411765 | 0.6923076923076923 | 0.72 | 0.6808510638297872 | 0.0005688667297363 | 0.0972669124603271 |
| SVM | 0.8518518518518519 | 0.8846153846153846 | 0.8214285714285714 | 0.8297872340425532 | 0.0274648666381835 | 0.0030479431152343 |
| XGBoost | 0.9122807017543859 | 1.0 | 0.8387096774193549 | 0.8936170212765957 | 0.241973876953125 | 0.0040738582611083984 |
| RoBERTa | 0.9230769230769231 | 0.9230769230769231 | 0.9230769230769231 | 0.9148936170212766 | 24968.250607967377 | 68.44634509086609 |
| Title | 📰 Project Page | Deployed Repository link | Tools Details | Notbook | Dataset | Medium | Diagram |
|---|---|---|---|---|---|---|---|
| 🌐1- Thesis | 1-2[-2] |
| Title | Published Date | Research Questions | Model performance metrics | Research Gap | Taking Notes | Input/Target Features | Journal name/ Category | Limitations | Future Directions |
|---|---|---|---|---|---|---|---|---|---|
| 🌐1- The development of a chatbot using Convolutional Neural Networks | 2022 | 1.How do different CNN architectures impact chatbot performance? | 1.Accuracy 2.Training Speed | Did author compared CNN chatbot to simpler models or rule-based systems? How well does it handle complex or open ended questions? | |||||
| 🌐2- Machine learning algorithms for teaching AI chat bots | 2021 | 1. Which machine learning algorithms are most successful in training AI chatbots for various tasks? | N/A | The paper does not cover various methods for evaluating the effectiveness of chatbot training algorithms. How can we measure a chatbot's ability to hold natural conversations, understand user intent, and generate appropriate responses? | Microservice architecture is used and the speed of message processing and preparation of responses by the chatbot will not change depending on the load on the server and the number of incoming messages. | ||||
| 🌐3- Personified Robotic Chatbot Based On Compositional Dialogues | 2022 | Research likely doesn't focus on specific questions but rather explores how compositional dialogues (where conversations are built from smaller elements) can be used to create a personified robotic chatbot. | N/A | 1. How effectively can the level of personality be measured in these chatbots? 2.Is user perception the only metric, or can objective measures be developed? | |||||
| 🌐4- Boosting the Accuracy of Optimization Chatbot by Random Forest with Halving Grid Search Hyperparameter Tuning | 2023 | 1.Can hyperparameter tuning with a Halving Grid Search method improve the accuracy of an optimization chatbot built using a Random Forest algorithm? | 1.Accuracy 2.Precision 3.Recall | The paper proposes three chatbot models: 1.One without hyperparameter tuning 2.One with hyperparameter tuning using Halving Grid Search 3.One with hyperparameter tuning and the best performing settings | |||||
| 🌐5- Developing a Chatbot using Machine Learning | 2019 | 1.Can machine learning algorithms improve the ability of a chatbot to understand natural language queries? 2. How does the choice of machine learning model (e.g., recurrent neural networks, decision trees) impact the performance of a chatbot? | 1. BLEU Score (BiLingual Evaluation Understudy) 2.Turing Test | This paper does not Investigate the impact of different visual design elements on user attention and engagement with the chatbot. | |||||
| 🌐6- Designing a Chatbot for Contemporary Education: A Systematic Literature Review | 2023 | What are the steps for designing an educational chatbot for contemporary education? | N/A | It focuses on the development of chatbots for education, not the impact on learners or educators. | |||||
| 🌐7- Research on the Design of Intelligent Chatbot Based on Deep Learning | 2021 | It is likely centered around improving the response generation of chatbots built with deep learning techniques. | N/A | Research gaps could exist in areas like sentiment analysis and generation of emotionally responsive dialogue. | Paper proposes an improved two-way GRU + Attention model based on the idea of mutual information, and examines the quality of the model from the final response effect. | ||||
| 🌐8- Question Answering Model Based Conversational Chatbot using BERT Model and Google Dialogflow | 2021 | N/A | N/A | The focus might be on question answering. Future research could explore integrating functionalities like sentiment analysis to tailor responses to user emotions or incorporating functionalities for completing tasks beyond just answering questions. | The focus of the paper seems to be on building and demonstrating the feasibility of a question-answering chatbot using BERT and Dialogflow. It describes the architecture and functionalities of the chatbot | ||||
| 🌐9- Chatbot : A Question Answering System for Student | 2021 | It suggests the research question that revolves around developing a chatbot system that effectively functions as a question answering system for students. | N/A | N/A | Paper discusses the design and development of such a chatbot, including the challenges of creating a system that can understand and answer student queries effectively. | ||||
| 🌐10-QAM: Question Answering System Based on Knowledge Graph in the Military | 2020 | How can a knowledge graph-based Question Answering System (QAM) be effectively designed to be used in the military domain? | N/A | Slice of words not included in the JIEBA will be divide, which cause that the following steps can’t accuracy judged. And some unclear words often led to the system failed to judge the right answer and return a wrong answer to the user. | Research used the tool of NEO4J to build the military KG as well python to construct QA system |
| Title | Defance Date/Published Date | Research Questions | Model performance metrics | Research Gap | Dataset | Notebook | Medium | Diagram |
|---|---|---|---|---|---|---|---|---|
| 🌐1- Thesis | 1-2-2 | Content 3 | ||||||
| 🌐2- Research Paper? | 1-2-3-4-5 | -5 |
| Title | Public_URL | Deployed Repository link | Tools Details | Notbook | Dataset | Medium | Diagram |
|---|---|---|---|---|---|---|---|
| 🌐1- Thesis | 1-2[-2] |
- 4 Chatbot Project with python
- Awesome Chatbot Projects
- The Super Duper NLP Repo
- Innovative-Chatbot-using-1-Dimensional-Convolutional-Layers
- 3XFake AI Technology: AI, ML, ANNs, Generative AI, LLMs, GPT-5, AGI
- Spam Detection using Machine Learning Methods
- Top AI Agents
- Learn to Train and Deploy a Real-Time Financial Advisor
- Self-RAG: Learning to Retrieve, Generate, and Critique through Self-Reflection
- Advanced RAG 08: Self-RAG
- Build a chatbot with always updated data sources using Pathway + LlamaIndex + Streamlit(Github)
- Practical NLP Code Chatbot-BERT(github)
- Anaconda and Vscode configuration for Python
- Get started with conda environments
- Getting Started with GitHub Pages
- How to deploy ML models painlessly
- Deploy Machine Learning models with FastAPI, Docker, and Heroku
- End to End Machine Learning project implementation (Part 1)
- Sentiment Analysis classifier with NLTK's VADER and Huggingface Roberta Transformers
- A Practical Guide to Transfer Learning using PyTorch
- Running Flask App On Colab With Ngrok| [ Latest Way ]
- Deploy Streamlit app on Google Colaboratory as public app | Ngrok | Python
- Building a Simple Chatbot with Python and Transformers
| Title | Published Date | Research Questions | Model Performance Metrics | Research Gap | Taking Notes | Input/Target Features | Journal name/ Category | Limitations | Future Directions |
|---|---|---|---|---|---|---|---|---|---|
| Predicting at-Risk Students at Different Percentages of Course Length for Early Intervention Using Machine Learning Models | IEEE Access. 2021 | To predict at-risk students at 0 to 100% of course length | "Accuracy at 20,40,60,80,100 was 0.59%, 0.79%, 0.84%, 0.88%, 0.90% and 0.91%" | - | - | ||||
| Development of Early Warning Systems to monitor e-learning progress | IOP Conference Series: Materials Science and Engineering 2021 | To predict whether students enrolled in a class will pass or fail and develop an early warning system to warn students and teachers about the possibility of students failing a course/class and suggest appropriate actions. | 95% Accuracy 4.9% Error | - | - | ||||
| Dropout early warning systems for high school students using machine learning | Children and Youth Services Review 2019 | To predict students at risk of dropping out using a Random forest classifier. | Accuracy=0.95 Sensitivity=0.85 Specificity=0.95 AUC of ROC = 0.97 | - | - | ||||
| Combining supervised and unsupervised machine learning algorithms to predict the learners’ learning styles | Second International Conference on Intelligent Computing in Data Sciences (ICDS 2018) | To predict the learning style of the students | Accuracy = 0.89 | - | - | ||||
| Integration of learning analytics into learner management system using machine learning | 2nd International Conference on Modern Educational Technology 2020 | To integrate Learning Analytics using Machine Learning into LMS | LG = 61% KNN = 60% DT = 63% | - | - | ||||
| Analysis and Prediction of Students’ Academic Performance Based on Educational Data Mining | IEEE Access 2022 | To combine analysis and prediction domains of EDM and analyze student performance using Clustering(KNN) and then using the output of clustering as labels in order to predict student future performance using classification (CNN). | 94.59%, 94.29% and 93.29% accuracy with the random hold-out split of 80-20 for the 3 data-sets A,B,C belonging to different courses. 91.22%, 95.92% and 93.90% when using 5-fold cross validation. | - | - |
| Title | Defance Date/Published Date | Research Questions | Model performance metrics | Research Gap | Dataset | Notebook | Medium | Diagram |
|---|---|---|---|---|---|---|---|---|
| 🌐1- Thesis | 1-2-2 | Content 3 | ||||||
| 🌐2- Research Paper? | 1-2-3-4-5 | -5 |
| f1 | precision | recall | accuracy | training_time | inference_time | |
|---|---|---|---|---|---|---|
| NB | 0.830188679245283 | 0.8461538461538461 | 0.8148148148148148 | 0.8085106382978723 | 0.0032272338867187 | 0.0007801055908203 |
| LR | 0.8518518518518519 | 0.8846153846153846 | 0.8214285714285714 | 0.8297872340425532 | 0.0356471538543701 | 0.0002150535583496 |
| KNN | 0.7058823529411765 | 0.6923076923076923 | 0.72 | 0.6808510638297872 | 0.0005688667297363 | 0.0972669124603271 |
| SVM | 0.8518518518518519 | 0.8846153846153846 | 0.8214285714285714 | 0.8297872340425532 | 0.0274648666381835 | 0.0030479431152343 |
| XGBoost | 0.9122807017543859 | 1.0 | 0.8387096774193549 | 0.8936170212765957 | 0.241973876953125 | 0.0040738582611083984 |
| RoBERTa | 0.9230769230769231 | 0.9230769230769231 | 0.9230769230769231 | 0.9148936170212766 | 24968.250607967377 | 68.44634509086609 |
| Title | Public_URL | Deployed Repository link | Tools Details | Notbook | Dataset | Medium | Diagram |
|---|---|---|---|---|---|---|---|
| 🌐1- Thesis | 1-2[-2] |
| Title | Published Date | Research Questions | Models performance | Research Gap | Taking Notes | Input/Target Features | Journal Category | Limitations | Future Directions |
|---|---|---|---|---|---|---|---|---|---|
| [1-Automatic Summarization of Russian Texts:Comparison of Extractive and Abstractive Methods/Computational Linguistics and Intellectual Technologies: Proceedings of the International Conference “Dialogue 2022” | June 15–18, 2022 | the methods under investigation have been ranked based on the ROUGE-N, ROUGE-L, BLEU, METEOR and BERTScore quality metrics and the salient features of summaries obtained by different methods have been revealed. | Gazeta= mBART= 31.55 13.54 28.22 ruT5-base =30.45 12.63 27.41, MLSUM=mBART 11.48 1.95 10.26 ,ruT5-base 12.35 1.86 11.22, XLSUM=mBART 26.47 10.95 22.67 ,ruT5-base 26.52 10.67 22.79 | for the first time, there has been carried a simultaneous comparison of extractive (TextRank and LexRank) and abstractive (mBART, ruGPT-3 and ruT5)summarization methods using three corpora of news articles: Gazeta, MLSUMand XL-Sum | results are compared graphically | ||||
| [2-DACSA: A large-scale Dataset for Automatic summarization of Catalan and Spanish newspaper Articles | 2022 | the construction of a corpus of Catalan and Spanish newspapers, the Dataset for Automatic summarization of Catalan and Spanish newspaper Articles (DACSA) corpus. | Average F1 scores of the models in the summarization task in Spanish=mBART= 30.66 12.08 23.13 23.89 71.07, mT5= 30.61 12.36 23.53 24.05, Average F1 scores of the models in the summarization task in Catalan.=mBART 27.46 11.04 21.13 22.01 , mT5 27.00 11.28 21.27 22.01 | The main objective of this work was to build a quality large-scale corpus that could be used to learn automatic summarization neural models for Catalan and Spanish. | |||||
| 3-Improving Abstractive Text Summarization with History Aggregation | 2019 | a novel aggregation mechanism to redistribute context states of text with collected history information. Then we equip the Transformer model with the aggregation mechanism. | model outperforms 1.01 ROUGE-1, 0.30 ROUGE-2 and 1.27 ROUGE-L scores on CNN/DailyMail dataset and 5.31 ROUGE-1, 4.56 ROUGE-2 and 5.19 ROUGE-L scores on our build Chinese news dataset compared to Transformer baseline model. | a novel aggregation mechanism to redistribute context states of text with collected history information. Then we equip the Transformer model with the aggregation mechanism. | |||||
| 4-Towards Zero-Shot Conditional Summarization with Adaptive Multi-Task Fine-Tuning | 2023 | An analysis of the role of 21 question answering, single- and multi-document summarization, causal reasoning, and argumentation tasks on zero-shot domain specific and general domain conditional summarization tasks. | T5= 0.43, 0.24, 1.70, 0.39, 0.61, BART=1.11, 0.37, 1.38, 0.46,0.62 | exploring the impact of multi-task fine-tuning (MTFT) on zero-shot conditional summarization for consumer health questions (MEDIQA, Savery et al., 2020) as well as topic-driven news article summarization | |||||
| 5-T5-Based Model for Abstractive Summarization: A Semi-Supervised Learning Approach with Consistency Loss Functions | 6/14/2023 | a novel semi-supervised learning method for abstractive summarization. To achieve this, a T5-based model to process texts and utilized an identity mapping constraint and a cycle consistency constraint to exploit the information contained in unlabeled data is employed. | BERT-base 63.83 51.29 59.76 41.45 , T5 50 with CL (ours) 53.13 41.03 50.85 33.95 , T5 250 with CL (ours) 59.41 47.93 56.16 38.91 | successful application of CycleGAN’s training process and loss functions to NLP tasks, particularly text summarization. |
| Title | Defance Date/Published Date | Research Questions | Model performance metrics | Research Gap | Dataset | Notebook | Medium | Diagram |
|---|---|---|---|---|---|---|---|---|
| 🌐1- Thesis | 1-2-2 | Content 3 | ||||||
| 🌐2- Research Paper? | 1-2-3-4-5 | -5 |
| Title | Public_URL | Deployed Repository link | Tools Details | Notbook | Dataset | Medium | Diagram |
|---|---|---|---|---|---|---|---|
| 🌐1- Thesis | 1-2[-2] |
- Advanced RAG 08: Self-RAG
- [Build a chatbot with always updated data sources using Pathway + LlamaIndex + Streamlit(Github](https://github.com/pathway-labs/realtime-indexer-qa-chat ref=blog.streamlit.io&fbclid=IwAR0HkaSIQTDCTL7Vw-V6sUMjR1L4vT9movAm3bCfX3ctqLTeALSPkmLMLRY)
- Practical NLP Code Chatbot-BERT(github)
- Anaconda and Vscode configuration for Python
- Get started with conda environments
- Getting Started with GitHub Pages
| Title | Published Date | Research Questions | Models performance | Research Gap | Taking Notes | Input/Target Features | Journal name/ Category | Limitations | Future Directions |
|---|---|---|---|---|---|---|---|---|---|
| 🌐1-Adaptive traffic signal control for real-world scenarios in agent based transport simulations | 2019 | N/A | |||||||
| 🌐2- What is machine learning? | 1-2-3-4-5 | -5 | |||||||
| 🌐3-Types of Machine Learning? | 1-2-3 | --- | |||||||
| 🌐4-Steps involved in Building a Machine Learning Model | 1 | --- | |||||||
| 🌐5-Best Free Resources to Learn Machine Learning | --- | -- |
| Title | Defance Date/Published Date | Research Questions | Model performance metrics | Research Gap | Dataset | Notebook | Medium | Diagram |
|---|---|---|---|---|---|---|---|---|
| 🌐1- Thesis | 1-2-2 | Content 3 | ||||||
| 🌐2- Research Paper? | 1-2-3-4-5 | -5 |
| Title | Public_URL | Deployed Repository link | Tools Details | Notbook | Dataset | Medium | Diagram |
|---|---|---|---|---|---|---|---|
| 🌐1- Thesis | 1-2[-2] |
| Title | Citation | Objective | Strengths | Weaknesses | ML Techniques | Models Performance | Research Gap | Taking Notes | Input/Target Features | Journal Category | Limitations | Future Directions |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 🌐 1. Phishing website detection based on effective machine learning approach | Harinahalli Lokesh, 2021 | Developing a model that predicts whether the website is a phishing site or a legitimate website. | Effective at detecting phishing with a focus on user training and software detection. Random Forest gives the highest accuracy of 96.87%. | Low importance features are also used in the model training. Confusion Matrix not provided. Just Accuracy Taken into consideration, no precision and recall etc. | One Class SVM, Linear SVC, K-Nearest Neighbor, Decision Tree Classifier, Random Forest | One Class SVM - 48.56% Accuracy, Linear SVC - 92.69%, K-Nearest Neighbor - 93.53%, Decision Tree Classifier - 96.05%, Random Forest - 96.87% | ||||||
| 🌐 2. Phishing web site detection using diverse machine learning algorithms | Zamir, 2020 | Using Neural Networks to predict whether a website is phishing or not. | Receives 97.4% Accuracy. It removes least important features by Recursive Feature Elimination technique. | This is using just 1055 rows of dataset that is less. This is taking 105.32 seconds to predict the website. | NB, KNN, SVM, RF, Bagging, NN | Naive Bayes - 72.67%, KNN - 94.2%, SVM - 93.1%, RF - 96.9%, Bagging - 95.1%, NN - 95.8% | ||||||
| 🌐 3. Phishing Detection Using Machine Learning Techniques | Shahrivari, 2020 | Building a best model using 12 ML models and selecting the best prediction model | High accuracy, detailed comparison across multiple ML techniques, considers recall, precision, and F-measure | Requires a large dataset, computationally intensive, feature selection is crucial | Logistic Regression, Decision Tree, Random Forest, AdaBoost, KNN, Neural Network, SVM, Gradient Boosting, XGBoost | Logistic Regression - 92.65%, Decision Tree - 96.60%, Random Forest - 97.26%, AdaBoost - 93.69%, KNN - 95.81%, Neural Network - 96.87%, SVM - 92.77%, Gradient Boosting - 94.86%, XGBoost - 98.23% | ||||||
| 🌐 4. Phishing URL detection using machine learning methods | Jha, 2023 | Building a strong prediction model using ML for detecting phishing and legitimate sites | Achieves high accuracy with LightGBM, effectively uses a limited feature set, quick prediction times | Used only 15 features, which may miss some important indicators | LightGBM, Random Forest, Decision Tree, Logistic Regression, SVM | LightGBM - 86%, Random Forest - 85.3%, Decision Tree - 85%, Logistic Regression - 84.2%, SVM - 83.5% | ||||||
| 🌐 5. Intelligent phishing detection scheme using deep learning algorithms | Adebowale, 2023 | Developing a real-time tool for detecting phishing websites | High true positive rate, handles large-scale datasets effectively, robust performance | Logistic regression may struggle with overfitting, linear assumptions, and scalability issues | ANN, Random Forest, SVM, Logistic Regression, MultinomialNB | ANN - 87.34%, Random Forest - 89.63%, SVM - 89.84%, Logistic Regression - 96.37%, MultinomialNB - 95.75% | ||||||
| 🌐 6. Intelligent phishing detection scheme using deep learning algorithms | Ahammad, 2022 | Developed a deep learning-based phishing detection system combining CNN and LSTM algorithms with 25s detection time | Receive 93.28% accuracy. Recall,precision and Fmeasure take into consideration. | This is taking 50 seconds to predict the webiste. Time-consuming for image processing methods. The model requires a significant amount of data for training and may be computationally expensive. | CNN,LSTM,IPDS | CNN - 92.55, LSTM - 92.79, IPDS - 93.28 |
| Title | Defance Date/Published Date | Research Questions | Model performance metrics | Research Gap | Dataset | Notebook | Medium | Diagram | To Do List |
|---|---|---|---|---|---|---|---|---|---|
| 🌐1- Thesis | 1-2-2 | Content 3 | |||||||
| 🌐2- Research Paper? | 1-2-3-4-5 | -5 |
| Title | Public_URL | Deployed Repository link | Tools Details | Notbook | Dataset | Medium | Diagram |
|---|---|---|---|---|---|---|---|
| 🌐1- Thesis | 1-2[-2] |
- 1. What is a Phishing Website Detection - Overview
- 2. Phishing URL Detection with ML
- 3. Phishing Website Detection by Machine Learning Techniques Implementation Guide
- 4. Phishing URL Detection with ML Dataset Exploration
- 5. PhiUSIIL Phishing URL (Website)
- 6. Phishtank Phishing webiste URL's dataset exploration
- 7. Datasets for phishing websites detection
- 8. Phishing Dataset by Canadian Institute for Cybersecurity
- 9. Awesome Machine Learning for Cyber Security
| Title | Published Date | Research Questions | Model performance metrics | Research Gap | Taking Notes | Input/Target Features | Journal name/ Category | Limitations | Future Directions |
|---|---|---|---|---|---|---|---|---|---|
| 🌐1- Identifying Interrelated Factors of Fatal and Injury Traffic Accidents Using Association Rules | 2023 | What are the Interrelated Factors of Fatal Injury | Apriori algorithm (Support=0.05, Confidence=0.70 and Lift >1) | NA | Introduction, Methodology and the description of confidence, support and lift | ||||
| 🌐2- Road traffic accidents analysis using association rule mining and descriptive analytics | 2023 | What are the characteristics of road traffic accidents | FP Growth Algorithm (Support=0.45, Confidence=0.95 and Lift >1) | NA | Introduction | ||||
| 🌐3- A Novel Approach to Avoid Road Traffic Accidents and Develop Safety Rules for Traffic Using Crash Prediction Model Technique | 2023 | How can Crash Prediction Models (CPMs) developed through machine learning approaches contribute to minimizing road traffic accidents and developing effective safety rules for traffic? | Random Forest achieves the highest values of accuracy and precision of around 60% | NA | Introduction, Related work | ||||
| 🌐4- A NOVEL ROAD TRAFFIC ACCIDENTS PREDICTION MODEL WITH RANDOM CLASSIFIER AFTER HYPER-PARAMETER TUNED USING GRIDSEARCHCV | 2023 | How can we effectively predict road traffic accidents using a novel prediction model incorporating Random Forest Classifier after hyper-parameter tuning with GridsearchCV? | Gradient Boosting Classifier (84.9 Accuracy) | NA | Future work | ||||
| 🌐5- Fatality Prediction for Motor Vehicle Collisions: Mining Big Data Using Deep Learning and Ensemble Methods | 2023 | How effective are deep learning and ensemble methods in predicting the fatality outcome of motor vehicle collisions using large-scale datasets? | Neural Network (75% Accuracy) | NA | Introduction |
| Title | Defance Date/Published Date | Research Questions | Model performance metrics | Research Gap | Dataset | Notebook | Medium | Diagram | To Do List |
|---|---|---|---|---|---|---|---|---|---|
| 🌐1- Thesis | 1-2-2 | Content 3 | |||||||
| 🌐2- Research Paper? | 1-2-3-4-5 | -5 |
| Title | Public_URL | Deployed Repository link | Tools Details | Notbook | Dataset | Medium | Diagram |
|---|---|---|---|---|---|---|---|
| 🌐1- Thesis | 1-2[-2] |
| Title | Published Date | Research Questions | Model performance metrics | Research Gap | Taking Notes | Input/Target Features | Journal name/ Category | Limitations | Future Directions |
|---|---|---|---|---|---|---|---|---|---|
| 🌐1- Identifying Interrelated Factors of Fatal and Injury Traffic Accidents Using Association Rules | 2023 | What are the Interrelated Factors of Fatal Injury | Apriori algorithm (Support=0.05, Confidence=0.70 and Lift >1) | NA | Introduction, Methodology and the description of confidence, support and lift | ||||
| 🌐2- Road traffic accidents analysis using association rule mining and descriptive analytics | 2023 | What are the characteristics of road traffic accidents | FP Growth Algorithm (Support=0.45, Confidence=0.95 and Lift >1) | NA | Introduction | ||||
| 🌐3- A Novel Approach to Avoid Road Traffic Accidents and Develop Safety Rules for Traffic Using Crash Prediction Model Technique | 2023 | How can Crash Prediction Models (CPMs) developed through machine learning approaches contribute to minimizing road traffic accidents and developing effective safety rules for traffic? | Random Forest achieves the highest values of accuracy and precision of around 60% | NA | Introduction, Related work | ||||
| 🌐4- A NOVEL ROAD TRAFFIC ACCIDENTS PREDICTION MODEL WITH RANDOM CLASSIFIER AFTER HYPER-PARAMETER TUNED USING GRIDSEARCHCV | 2023 | How can we effectively predict road traffic accidents using a novel prediction model incorporating Random Forest Classifier after hyper-parameter tuning with GridsearchCV? | Gradient Boosting Classifier (84.9 Accuracy) | NA | Future work | ||||
| 🌐5- Fatality Prediction for Motor Vehicle Collisions: Mining Big Data Using Deep Learning and Ensemble Methods | 2023 | How effective are deep learning and ensemble methods in predicting the fatality outcome of motor vehicle collisions using large-scale datasets? | Neural Network (75% Accuracy) | NA | Introduction |
| Title | Defance Date/Published Date | Research Questions | Model performance metrics | Research Gap | Dataset | Notebook | Medium | Diagram | To Do List |
|---|---|---|---|---|---|---|---|---|---|
| 🌐1- Thesis | 1-2-2 | Content 3 | |||||||
| 🌐2- Research Paper? | 1-2-3-4-5 | -5 |
| Title | Public_URL | Deployed Repository link | Tools Details | Notbook | Dataset | Medium | Diagram |
|---|---|---|---|---|---|---|---|
| 🌐1- Thesis | 1-2[-2] |
| Title | Published Date | Research Questions | Model performance metrics | Research Gap | Taking Notes | Input/Target Features | Journal name/ Category | Limitations | Future Directions |
|---|---|---|---|---|---|---|---|---|---|
| 🌐1- Identifying Interrelated Factors of Fatal and Injury Traffic Accidents Using Association Rules | 2023 | What are the Interrelated Factors of Fatal Injury | Apriori algorithm (Support=0.05, Confidence=0.70 and Lift >1) | NA | Introduction, Methodology and the description of confidence, support and lift | ||||
| 🌐2- Road traffic accidents analysis using association rule mining and descriptive analytics | 2023 | What are the characteristics of road traffic accidents | FP Growth Algorithm (Support=0.45, Confidence=0.95 and Lift >1) | NA | Introduction | ||||
| 🌐3- A Novel Approach to Avoid Road Traffic Accidents and Develop Safety Rules for Traffic Using Crash Prediction Model Technique | 2023 | How can Crash Prediction Models (CPMs) developed through machine learning approaches contribute to minimizing road traffic accidents and developing effective safety rules for traffic? | Random Forest achieves the highest values of accuracy and precision of around 60% | NA | Introduction, Related work | ||||
| 🌐4- A NOVEL ROAD TRAFFIC ACCIDENTS PREDICTION MODEL WITH RANDOM CLASSIFIER AFTER HYPER-PARAMETER TUNED USING GRIDSEARCHCV | 2023 | How can we effectively predict road traffic accidents using a novel prediction model incorporating Random Forest Classifier after hyper-parameter tuning with GridsearchCV? | Gradient Boosting Classifier (84.9 Accuracy) | NA | Future work | ||||
| 🌐5- Fatality Prediction for Motor Vehicle Collisions: Mining Big Data Using Deep Learning and Ensemble Methods | 2023 | How effective are deep learning and ensemble methods in predicting the fatality outcome of motor vehicle collisions using large-scale datasets? | Neural Network (75% Accuracy) | NA | Introduction |
| Title | Defance Date/Published Date | Research Questions | Model performance metrics | Research Gap | Dataset | Notebook | Medium | Diagram | To Do List |
|---|---|---|---|---|---|---|---|---|---|
| 🌐1- Thesis | 1-2-2 | Content 3 | |||||||
| 🌐2- Research Paper? | 1-2-3-4-5 | -5 |
| Title | Public_URL | Deployed Repository link | Tools Details | Notbook | Dataset | Medium | Diagram |
|---|---|---|---|---|---|---|---|
| 🌐1- Thesis | 1-2[-2] |
| Title/Journal Name | Published Date | Research Questions | Model performance metrics | Research Gap | Taking Notes |
|---|---|---|---|---|---|
| 🌐1- Why we used AI | 1-2-2 | Content 3 | |||
| 🌐2- What is machine learning? | 1-2-3-4-5 | -5 | |||
| 🌐3-Types of Machine Learning? | 1-2-3 | --- | |||
| 🌐4-Steps involved in Building a Machine Learning Model | 1 | --- | |||
| 🌐5-Best Free Resources to Learn Machine Learning | --- | -- |
| Title | Defance Date/Published Date | Research Questions | Model performance metrics | Research Gap | Dataset | Notebook | Medium | Diagram |
|---|---|---|---|---|---|---|---|---|
| 🌐1- Thesis | 1-2-2 | Content 3 | ||||||
| 🌐2- Research Paper? | 1-2-3-4-5 | -5 |
| Title | Public_URL | Deployed Repository link | Tools Details | Notbook | Dataset | Medium | Diagram |
|---|---|---|---|---|---|---|---|
| 🌐1- Thesis | 1-2[-2] |
| Title | Public URL | Deployed Repository link | Tools Details | Notbook | Dataset | Medium | Diagram |
|---|---|---|---|---|---|---|---|
| 🌐1- Thesis | 1-2[-2] |
| Style Name | Journal Articles(Reference) | Books(reference) | website (reference) | Journal Articles(Citation) | Books(cite) |
|---|---|---|---|---|---|
| 🌐1- The development of a chatbot using Convolutional Neural Networks | 2022 | 1.How do different CNN architectures impact chatbot performance? | 1.Accuracy 2.Training Speed | Did author compared CNN chatbot to simpler models or rule-based systems? How well does it handle complex or open ended questions? | |
| 🌐2- Machine learning algorithms for teaching AI chat bots | 2021 | 1. Which machine learning algorithms are most successful in training AI chatbots for various tasks? | N/A | The paper does not cover various methods for evaluating the effectiveness of chatbot training algorithms. How can we measure a chatbot's ability to hold natural conversations, understand user intent, and generate appropriate responses? | Microservice architecture is used and the speed of message processing and preparation of responses by the chatbot will not change depending on the load on the server and the number of incoming messages. |
-
Fork the repository
-
Clone your forked repository using terminal or gitbash.
-
Make changes to the cloned repository
-
Add, Commit and Push
-
Then in Github, in your cloned repository find the option to make a pull request
print("Start contributing for Research-Work")
-
Make sure you do not copy codes from external sources because that work will not be considered. Plagiarism is strictly not allowed.
-
Plagiarism rules: For synopsis: 15% is acceptable as HEC rule and individuals source less than 2%
-
If you want to contribute the algorithm, it's preferrable that you create a new issue before making a PR and link your PR to that issue.
-
If you have modified/added code work, make sure the code compiles before submitting.
-
Strictly use snake_case (underscore_separated) in your file_name and push it in correct folder.
-
Step 1- Internal Department synopsis Presention, Required Documents( Presention PPT)
-
Step 2- Synopsis Presention For GRC, Required Documents(PPT,Synopsis File,AnnexB,AnnexA,Turnitin Report(15% is acceptable and individuals source less than 2%)).
-
Step 3 - Sent Thesis for Rreviewer , Required Documents(final thesis,Plagiarism report (15% is acceptable and individuals source less than 2%), Ssynopsis notification, suggested as external examiners,Annex F)
-
Step 4 - Response To Extrnal Reviewer,Plagiarism report,Compliance COMPLIANCE REPORT
-
Step 5-🎓 Final thesis defense
-
Step 6-🎓 Submit four Hard copies of thesis with signature of student and supervisory committee
We would love your help in making this repository even better! If you want to join research work then please contact.
Together, let's make this the best AI learning hub website! 🚀
Thanks goes to these Wonderful People. Contributions of any kind are welcome!🚀