OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
AI Data & Analytics

Mams For Absa

Mams For Absa is a research dataset and Python code repository for aspect-based sentiment analysis. Its MAMS challenge data includes sentences with multiple aspects carrying different sentiment polarities, with versions for aspect-term and aspect-category analysis; accompanying scripts support preprocessing, training and testing.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Mams For Absa — OpenViST project profile in AI Data & Analytics.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use it when you need a research dataset and local code workflow for aspect-based sentiment analysis, especially for sentences containing multiple aspects with different polarities.

Who it is for

  • NLP researchers benchmarking aspect-term sentiment analysis.
  • Practitioners evaluating aspect-category sentiment classifiers.
  • Developers building local Python training and testing workflows for ABSA.

Documented uses

  • Evaluate how a model identifies sentiment for multiple aspects within a sentence.
  • Benchmark aspect-term or aspect-category sentiment analysis using the corresponding dataset variant.
  • Preprocess data and train and test a model with the repository scripts.
Before you choose

Requirements & limits

The GitHub repository is public and identifies an Apache-2.0 license.

  • Choose term mode with an ATSA dataset for aspect-term sentiment analysis; choose category mode with an ACSA dataset for aspect-category analysis.
  • The documented workflow depends on listed older package versions and a separately obtained pretrained GloVe file.
  • The repository provides code and data rather than a hosted analysis interface.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Prepare and run the documented ABSA workflow

Configure an ATSA or ACSA task and run the repository's preprocessing, training and testing scripts.

Before you start

  • Python environment with the README-listed versions of PyTorch, spaCy, pytorch-pretrained-bert, AdaBound, PyYAML, NumPy, scikit-learn and SciPy.
  • Pretrained GloVe file glove.840B.300d.txt placed in ./data.
  • An ATSA dataset for term mode or an ACSA dataset for category mode.

Steps

  1. Edit config.py to select the task, model and hyperparameters.

  2. Set mode to term and point base_path to an ATSA dataset, or set mode to category and point base_path to an ACSA dataset.

  3. Run python preprocess.py.

  4. Run python train.py, then python test.py.

Expected outcome

The documented scripts preprocess the selected dataset, train a model and run testing for the configured task.

Guide limitations

  • The README does not specify dataset acquisition or detailed configuration values.
  • This workflow requires the specified local dependencies and pretrained GloVe file.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
17 October 2026
Original website title
siat-nlp/MAMS-for-ABSA · Official homepage title · retained in its source language
How we review sourcesSuggest a correction