siat-nlp/MAMS-for-ABSA siat-nlp/MAMS-for-ABSA

  • Stars
    star
    254
  • Rank 159,356 (Top 4 %)
  • Language
    Python
  • License
    Apache License 2.0
  • Created about 5 years ago
  • Updated almost 5 years ago
Twitter logo Share LinkedIn logo Share Facebook logo Share
siat-nlp/MAMS-for-ABSA
github

siat-nlp

Reviews

There are no reviews yet. Be the first to send feedback to the community and the maintainers!

Repository Details

A Multi-Aspect Multi-Sentiment Dataset for aspect-based sentiment analysis.

MAMS-for-ABSA

This repository contains the data and code for the paper "A Challenge Dataset and Effective Models for Aspect-Based Sentiment Analysis", EMNLP-IJCNLP 2019, [paper].

MAMS

MAMS is a challenge dataset for aspect-based sentiment analysis (ABSA), in which each sentences contain at least two aspects with different sentiment polarities. MAMS dataset contains two versions: one for aspect-term sentiment analysis (ATSA) and one for aspect-category sentiment analysis (ACSA).

Requirements

pytorch==1.1.0
spacy==2.1.8
pytorch-pretrained-bert==0.6.2
adabound==0.0.5
pyyaml==5.1.2
numpy==1.17.2
scikit-learn==0.21.3
scipy==1.3.1

Quick Start

Put the pretrained GloVe(http://nlp.stanford.edu/data/wordvecs/glove.840B.300d.zip) file glove.840B.300d.txt in folder ./data. Modify config.py to select task, model and hyper-parameters. When mode is set to term, base_path should point to an ATSA dataset. When mode is set to category, base_path should point to an ACSA dataset.

Preprocessing

python preprocess.py

Train

python train.py

Test

python test.py

Acknowledgement

The BERT model pretrained by huggingface(https://github.com/huggingface/pytorch-transformers) is used in our experiments.

More Repositories

1

GALAXY

Official repository of the AAAI'2022 paper "GALAXY: A Generative Pre-trained Model for Task-Oriented Dialog with Semi-Supervised Learning and Explicit Policy Injection"
Python
106
star
2

HanFei

ๅ›ฝๅ†…้ฆ–ไธชๅ…จๅ‚ๆ•ฐ่ฎญ็ปƒ็š„ๆณ•ๅพ‹ๅคงๆจกๅž‹ HanFei-1.0 (้Ÿฉ้ž)
Python
81
star
3

TransDG

Source code for the paper "Improving Knowledge-aware Dialogue Generation via Knowledge Base Question Answering".
Python
64
star
4

CpRec

A Generic Network Compression Framework for Sequential Recommender Systems
Python
19
star
5

TTOS

Official repository of the EMNLP'2020 paper "Amalgamating Knowledge from Two Teachers for Task-oriented Dialogue System with Adversarial Training"
Python
16
star
6

transformer-pytorch

A PyTorch implementation of transformer for text generation.
Python
15
star
7

DDMN

Code and data for the paper "Dual Dynamic Memory Network for End-to-End Multi-turn Task-oriented Dialog Systems".
Python
13
star
8

NLP-docs

Docs of NLP/deep Learning/machine learning, etc. https://siat-nlp.github.io/docs
HTML
11
star
9

MGMA

Official repository of the KBS'2021 paper "Multi-Goal Multi-agent Learning for Task-oriented Dialogue with Two-to-One Teacher-Student Learning"
Python
7
star
10

dialogue-models

This repository contains reconstructed code for dialogue models.
Python
7
star
11

IPSRec

Code and data for the SIGIR'2022 paper "Item-Provider Co-learning for Sequential Recommendation" (IPSRec)
Python
6
star
12

seq2seq-pytorch

This repository contains RNN, CNN, Transformer based Seq2Seq implementation.
Python
5
star
13

Multi-task-Learning-for-Answer-Selection

A Multi-Task Learning Approach for Answer Selection: A Study and a Chinese Law Dataset
Python
3
star
14

Knowledge-driven-dialogue

Knowledge-driven-dialogue
Python
2
star
15

GPSP-for-Heterogeneous-Network-Embedding

GPSP: Graph Partition and Space Projection based Approach for Heterogeneous Network Embedding
Python
1
star
16

Learning-Document-Embeddings-with-Crossword-Prediction

Learning Document Embeddings with Crossword Prediction
Python
1
star
17

HSEMEC-code-data

Python
1
star