{ "fetched_at": "2026-10-19T09:23:16Z", "source_url": "https://www.kaggle.com/competitions/expertclass2/data", "pages": [ { "name": "data-description", "content": "\nThe dataset consists of *48x48* pixels grayscale of faces. The task is to categorize each face based on the emotion shown in the facial expression into one of seven categories:\n\n- 0: Angry\n- 1: Disgust\n- 2: Fear\n- 3: Happy\n- 4: Sad\n- 5: Surprise\n- 6: Neutral\n\n\n**File descriptions**\n\nWe're providing three CSV files. The `train_data.csv` contains the data to train your model and the `test_data.csv` has the images you're going to use to test. To submit a solution, you have to generate a file similar to the `example_submission.csv`.\n\n\nThe CSVs files contain two columns: *emotion* and *pixels*. \nThe *emotion* column has a numeric code ranging from 0 to 6 from the emotion in the image.\nThe *pixels* column has a string containing the image\n\nFor examples of how to handle the data, access the notebook [Getting started with the data](https://www.kaggle.com/hugozaniniabi/getting-started-with-the-data).\n\nYou can also download these data files via ABI SharePoint [Link](https://anheuserbuschinbev.sharepoint.com/:f:/s/AnalyticsAcademy/EkitLIW3hVtHm4eUWmyvVywBk3GO4ZHGJfFVsiTxzMOSYw?e=MSHeMJ)", "isPublished": false, "hasIsPublished": false, "postTitle": "", "hasPostTitle": false, "mimeType": "", "hasMimeType": false }, { "name": "Description", "content": "# Face Emotions Classification\n\n### Competition Description\nAt ABI we're always concerned about the customer experience and trying to understand their behavior to provide the best beverages in the market. However, only words do not match emotions and what a person is actually feeling.\n\nWhat if we create Machine Learning models to understand patterns in images and say if a customer is happy or not while it's drinking one of our beverages?\n\nThat's the goal of this challenge!! We are providing a dataset of 35K images of people in many situations and their feelings. **Are you able to create a model to understand what's their face is telling you?**\n\nGood Luck 🍀\n\n---\n### Participant Eligibility\n**To be eligible, you MUST be an ABI employee and have a valid corporate email address**\n\n\n---\n### Prize\nOne Winner - $100 (USD) Visa e-Gift Card \n\n---\n### Practice Skills\n- Computer vision fundamentals including simple neural networks\n- Classification methods such as SVM and K-nearest neighbors\n\n---\n### Hosts\n\n**Analytics Academy**\n[Tina Gui](mailto:tina.gui@ab-inbev.com?subject=[Kaggle%20Competition])\n[Bruno Raposo](mailto:bruno.raposo@ab-inbev.com?subject=[Kaggle%20Competition])\n[Gopakumar R](mailto:gopakumar.r@ab-inbev.com?subject=[Kaggle%20Competition])\n\n\n**AmbevTech**\n[Hugo Zanini](mailto:hugozanini@ambev.com.br?subject=[Kaggle%20Competition])\n\n\n\n---\n### Acknowledgements\nThe images were gathered from many datasets and it's a mix of the following works:\n- [The Extended Yale Face Database B](http://vision.ucsd.edu/~iskwak/ExtYaleDatabase/ExtYaleB.html)\n- [Labeled Faces In The Wild](http://vis-www.cs.umass.edu/lfw/)\n- [The Japanese Female Facial Expression (JAFFE) Database](https://zenodo.org/record/3451524#.YHeSy-hKhPY)\n- [Indian Movie Face database (IMFDB)](http://cvit.iiit.ac.in/projects/IMFDB/)\n\n\n---\n### References\n[0] https://www.verywellmind.com/understanding-emotions-through-facial-expressions-3024851\n\n[1] Cover image: https://www.stocksy.com/463065/selfie-friends-in-pub", "isPublished": false, "hasIsPublished": false, "postTitle": "", "hasPostTitle": false, "mimeType": "", "hasMimeType": false }, { "name": "Evaluation", "content": "## Evaluation\n\nThe evaluation metric for this competition is categorization accuracy. It is calculated by the fraction of predictions our model got right. Formally, accuracy has the following definition:\n\n![enter image description here](https://tutorialspoint.dev/image/Confusion_Matrix2_2.png)\n\n\nTP: True Positive\nTN: True Negative\nFP: False Positive\nFN: False Negative\n\n## Submission Format\nYou should submit a file containing the predictions for the `test_data.csv` images. The submission file should contain two columns: ID and Emotion\nFor more details, see the [example_submission.csv](https://www.kaggle.com/c/expertclass2/data?select=example_submission.csv) file.\n\n## Leaderboard\n\nThe `test_data.csv` is pre-divided into Part A and Part B to generate a public and a private leaderboard. Part A will be used to generate the accuracy of the model (displayed public) and Part B to check possible overfittings. Due to that, the results may vary if the solution is not generalizing well, but you will be notified if it's happened. \n \nEach part will have a weight of 50% in the result. The model must have good accuracy and the ability to predict unknown data.\n\nFinal score = (public leaderboard score * 50%) + (private leadership score * 50%)\n", "isPublished": false, "hasIsPublished": false, "postTitle": "", "hasPostTitle": false, "mimeType": "", "hasMimeType": false }, { "name": "rules", "content": " This competition is designed for ABI employees only.\n\n - Please register a new Kaggle account with ABI email address.\n - You're encouraged to enter as a pair or team. You can only be in one team.\n - Don't cheat!\n - This data set is publicly available on the internet, including test data. Do not use this test data in training!\n - We're going to check all the code used to train the models. The results should be reproducible.\n - Apply yourself!\n - Have fun!", "isPublished": false, "hasIsPublished": false, "postTitle": "", "hasPostTitle": false, "mimeType": "", "hasMimeType": false }, { "name": "Winners", "content": "## Winners Announcement\n\nCongratulations to the Winning Team - \"**Brasil com S**\". \n\nThe team is composed by:\n- Adauto Pereira de Almeida\n- Fabricio Talarico\n- Guilherme Rodrigues Barbazza\n- Lucas Neto Nakadaira\n\n", "isPublished": false, "hasIsPublished": false, "postTitle": "", "hasPostTitle": false, "mimeType": "", "hasMimeType": false } ] }