Use case on the amazon-commerce-reviews dataset

Dataset amazon-commerce-reviews

Machine Learning Task: Multiclass classification

This is an amazon-commerce-reviews. Datasets are derived from the customer's reviews on Amazon Commerce Website for authorship identification. Most previous studies conducted identification experiments for two to ten authors. But in the online context, reviews to be identified usually have more potential authors, and normally classification algorithms are not adapted to a large number of target classes. To examine the robustness of classification algorithms, the authors of this database identify 50 of the most active users (represented by a unique ID and username) who frequently posted reviews in these newsgroups. The number of reviews we collected for each user is 30.

Available at OpenML:

Category: Marketing

# Rows: 1,500 # Columns: 10,000

Target: Class


Numeric: V1, V2, V3, V4, V5, V6, V7, V8, V9, V10, V11, V12, V13, V14, V15, V16, V17, V18, V19, V20, ...

Machine Learning Use Case Marketing

Cross-Entropy Loss (LOGLOSS)

Amazon Commerce Reviews Logloss

Accuracy (ACC)

Amazon Commerce Reviews Acc

Balanced Accuracy (BALACC)

Amazon Commerce Reviews Balacc

« Back to Machine Learning Algorithms Comparison