{ "fetched_at": "2026-10-19T19:33:19Z", "source_url": "https://www.kaggle.com/competitions/cmv6-ml1/data", "pages": [ { "name": "rules", "content": "# Rules:\n\n## 1. every participant is required to fill out [this google form ](https://forms.gle/pmtaiuFr4snsGyWx9)before 13 March 2020 (IST) or making a submission to this contest(whatever is earlier). \n\n\n## 2. every participant is required to submit and make his/her solution public as kaggle kernels, on contest end, which should have runtime less than 6hr (using GPU/TPU). Kernel/notebook should be working and producing same result as submitted during contest.\n\n\n## 3. Dataset is created with a lot of effort, so Use of this Dataset is limited to this contest only.\n\n\n## 4. one participant, one Kaggle user ID policy applies with this contest too.\n\n\n## 5. Plagiarism is auto detected by Kaggle, same submission by multiple user will lead to dis qualification.\n\n\n## 6. Use of Pretrained Models: you are free to use any Pre trained model available publcly unless it is violating any of the rules of this contest.\n\n\n## 7. Have fun, Don't cheat, stress your potentials.\n ", "isPublished": false, "hasIsPublished": false, "postTitle": "", "hasPostTitle": false, "mimeType": "", "hasMimeType": false }, { "name": "data-description", "content": "# Train Data:\ntrainData.csv it contains three column\n\n1. id: unique id for each train data points.\n\n2. sentence: it is Hindi news headlines which needed to be classified wehter it is clickbait(1) or not clickbait(1)\n\n3. label : label corrosponding to each datapoints and id , 1(clickbait), 0(non clickbait)\n\n# Test Data\n\ntestData.csv\n\n1. id: unique id for each training data points.\n\n2. sentence: it is Hindi news headlines which needed to be classified whehter it is clickbait(1) or not clickbait(1)\n\neach test datapoints needs to be classified and results (0 or 1) needs to be submitted in a csv file with 'id' and its corresponding labels.\n\n#sample submission\nsample_submission.csv : exact format for submission file\n\nsubmission file should have two column:\n\n1. 'id': all unique id in test data\n2. 'label': whether clickbait(1) or not (0) for all data points in test data corresponding to each 'id'.\n\nPS: please take care to submit result for each and every data points in test data, otherwise will lead to submission error.", "isPublished": false, "hasIsPublished": false, "postTitle": "", "hasPostTitle": false, "mimeType": "", "hasMimeType": false }, { "name": "Description", "content": "*Hello ISMites*,\nThrough this contest, lets us indulge in solving a real life problem, which is affecting a lot of internet users. Clearly this is a NLP task, but lets change it a bit, we will apply NLP to **Hindi** Language this time. and lets see how much we could Improve NLP for Hindi language and NLP culture in ISM. \n\nA kind request to the participant would be treating this contest more as a **learning opportunity**, Please do start discussion in 'Discussion' section. If you have good tutorial you want to share, or have a idea you want to discuss or want feedback, please start a discussion for the same. \n\n*I would be regularly posting baseline solution, starter codes and tutorial for newbies.*\n\n**Please take a look at the Rules section.**\n\n# ClickBait Detection For Hindi Language\n## what is ClickBait?\n [Clickbait](https://en.wikipedia.org/wiki/Clickbait), a form of false advertisement, uses hyperlink text or a thumbnail link that is designed to attract attention and to entice users to follow that link and read, view, or listen to the linked piece of online content, with a defining characteristic of being deceptive, typically sensationalized or misleading.\n\n## why it is necessary?\nClick-bait waste a lot of productive time of people. These headlines are written in a very catchy manner such that people are tempted to click these links and they don’t contain any relevant information, hence making it necessary to warn a user about click-bait.\n\n## Task\n\nTrain A binary text classifier to detect clickbait in Hindi Languages. Dataset provided is headlines of news corpus in Hindi Language, with label of '0' and '1'.\n - '1' Indicating clickbait\n - '0' Indicating non- clickbait.\n\n## Evaluation\n For evauation we will be using F1-score matric. you would be scored initially on Public data which will be 35% of original test data. Final result would be on total test data and will be made public after contest ends.\n\n## code submission\n ***each participant is required to submit working notebook as kaggle kernel, which should have runtime less than 6 hrs.*** you can use pretrained models available publicly, fine tune them, but all main script (i.e model declaration, weight loader, inference notebook) should be submitted and all of them should be working and producing same result", "isPublished": false, "hasIsPublished": false, "postTitle": "", "hasPostTitle": false, "mimeType": "", "hasMimeType": false }, { "name": "Evaluation", "content": "\n
\n

The evaluation metric for this competition is Mean F1-Score. The F1 score, commonly used in information retrieval, measures accuracy using the statistics precision p and recall r. Precision is the ratio of true positives (tp) to all predicted positives (tp + fp). Recall is the ratio of true positives to all actual positives (tp + fn). The F1 score is given by:

\n

\\[ F1 = 2\\frac{p \\cdot r}{p+r}\\ \\ \\mathrm{where}\\ \\ p = \\frac{tp}{tp+fp},\\ \\ r = \\frac{tp}{tp+fn} \\]

\n

The F1 metric weights recall and precision equally, and a good retrieval algorithm will maximize both precision and recall simultaneously. Thus, moderately good performance on both will be favored over extremely good performance on one and poor performance on the other.

\n", "isPublished": false, "hasIsPublished": false, "postTitle": "", "hasPostTitle": false, "mimeType": "", "hasMimeType": false }, { "name": "Timeline", "content": "**Last Date of submission: 23:59, 16 March 2020.**", "isPublished": false, "hasIsPublished": false, "postTitle": "", "hasPostTitle": false, "mimeType": "", "hasMimeType": false }, { "name": "Prizes", "content": "## Prize for contest winner\nTo be declared soon.\n\n\n## Prize for a best kernel and discussion\n\nParticipant posting best relevant kernel/tutorial or starting best discussion will get a treat from Final year organizers.", "isPublished": false, "hasIsPublished": false, "postTitle": "", "hasPostTitle": false, "mimeType": "", "hasMimeType": false }, { "name": "Tutorials", "content": "# Beginners\n Refer to this tuorial for getting started with NLP. -> [Tutorial 1](https://www.analyticsvidhya.com/blog/2017/01/ultimate-guide-to-understand-implement-natural-language-processing-codes-in-python/)\n\n In case of doubt related to this tutorial you can start a thread in the discussion section.\n\nParticipants are highly encouraged to start a discussion.\n\n### Overview of Major steps in NLP:\n( these steps are applied for all languages)\n1. **Text pre processing**: cleaning, noise removal, spelling correction, punctuation handling, etc .\n2. **Language Modeling:** since computer doesn't understand text, we need to convert the text into some numerical equivalent things like, word embedding, tf-idf and using other method of representation. Tokenization, vocabulary creation are other part of this step .\n3. **Model Training:** In this step, we train a ML/Neural network (CNN, RNN, Transformer, etc) model to achieve our task.\n\n# NLP for hindi Language\n\nrefer this tutorial for NLP for Hindi languages: [Tutorial 2](https://www.analyticsvidhya.com/blog/2020/01/3-important-nlp-libraries-indian-languages-python/) \n\nIn case of doubt related to this tutorial you can start a thread in the discussion section.\n", "isPublished": false, "hasIsPublished": false, "postTitle": "", "hasPostTitle": false, "mimeType": "", "hasMimeType": false } ] }