100 datasets found

The Taobao User Behavior dataset is a subset of user behaviors on Taobao collected within 9 days and consists of more than 70 million samples and 1 million users.

The dataset used in this paper is Yahoo!Movies dataset provided by Yahoo!Research Alliance Webscope program.

FourSquare shopping places dataset from 5 cities in Indonesia. It consists of 1) 176 shopping places data, 2) 3844 visitors' reviews, and 3) 14309 users data.

A dataset of user job application history and active job posting records.

The dataset is a two-mode tensor of user and movie ratings, of size 6, 040 × 3, 706.

The dataset is a continuous tensor representing the three-way interactions (user, action, file), of size 3, 000 × 150 × 30, 000.

The Beauty dataset is a product review dataset crawled from Amazon. The data is split into separate datasets by the top-level product category.

The dataset used in the paper for continuous-time sequential recommendation task

The dataset is used for sequential recommendation tasks, and it contains user-item interaction history.

The MQ2008 dataset is a large-scale web search dataset, containing approximately 100,000 query-url pairs with relevance scores.

The MSLR-WEB10K dataset is a large-scale web search dataset, containing approximately 235,000 query-url pairs with relevance scores.

The dataset is used for sequential recommendation tasks, and it contains user-item interaction history.

The MovieLens-IMDB dataset is a collection of user ratings for movies, with each rating indicating the user's preference for the movie.

The dataset used in the paper is a Netflix dataset, which is a large-scale matrix factorization problem.

Bing-News is a dataset containing 1,025,192 pieces of implicit feedback collected from the server logs of Bing News.

The Walmart.com dataset consists of ~100,000 shopping cart snapshots with the customers' continual shopping records.

The Instacart dataset consists of ~50,000 grocery products, with the shopping records of ~200 thousand users and 3 million orders.

Precise user modeling is critical for online personalized recommen-der services. Generally, users’ interests are diverse and are not limited to a single aspect, which is...

Amazon book data set contains book reviews and metadata from Amazon. Following previous work, we regard reviews as positive samples and randomly select products not rated by a...

The production data set collected from Alibaba display advertising platform. We randomly select about 1% of 1-day samples for training and 0.1% of the next-day samples for...