Complementary Teaching Material
Our teaching materials are meticulously crafted to provide researchers and students with practical insights and hands-on experience in navigating and using datasets for research in business, economics, and social sciences. Whether you are a beginner eager to understand the basics of data handling or an advanced data scientist seeking to refine your skillset, our educational content supports your learning journey at every step. Dive into our comprehensive teaching materials to discover how to leverage the full potential of our platform’s datasets.
Sentiment Analysis and Predictive Modelling: Rotten Tomatoes Reviews
The Rotten Tomatoes Movie Review Dataset comprises a vast compilation of critical appraisals and audience feedback across a wide range of movies, detailed through numerical ratings and textual reviews. In our tutorial, we guide participants to apply sentiment evaluation and predictive modeling to uncover underlying patterns in movie ratings and review sentiments.
Downloads
Computer Vision: CelebFaces Attributes Image Dataset
The CelebFaces Attributes Dataset (CelebA) is a comprehensive resource for applying advanced machine learning and computer vision methodologies in facial attribute recognition. Containing over 200,000 images, each annotated with 40 attributes alongside five key landmark locations, CelebA offers a large volume of data for academic and practical exploration in image processing. In this tutorial, participants are guided to implement and compare various machine learning techniques, from conventional logistic regression to advanced neural networks, to identify and predict image attributes effectively.
Downloads
Prediction Competition: Huawei DIGIX Advertisement CTR
Advertisement click-through-rate (CTR) prediction is the key problem in the area of defining impactful advertising campaigns. Increasing the accuracy of advertisement CTR prediction is critical to improve the effectiveness of precision marketing. In this competition, big advertising datasets have been released that are anonymized. Based on the datasets, contestants were required to build Advertisement CTR prediction models. Can you set up a competitive prediction model as well?
The datasets contain the advertising behavior data collected from seven consecutive days, including a training dataset and a testing dataset.