Light

contextscout / ned-graphs Goto Github PK

View Code? Open in Web Editor NEW

37.0 8.0 16.0 366.2 MB

Python 100.00%

ned-graphs's Introduction

Code and Dataset for Named Entity Disambiguation using Deep Learning on Graphs

This repository contains the code and dataset for the paper "Named Entity Disambiguation using Deep Learning on Graphs". The full paper can be found here.

Installation

The main requirements are installed with:

virtualenv --python=/usr/bin/python3 .env
source .env/bin/activate
pip install -r requirements.txt

Download the glove vector files

cd data
wget http://nlp.stanford.edu/data/glove.840B.300d.zip
unzip glove.840B.300d.zip
echo "2196017 300" | cat - glove.840B.300d.txt > glove_2.2M.txt
cd ..

One must also unzip the mapping of the wikidata nodes

cd data
bunzip2 x*
cat x* > wikidata_items.csv
cd ..

Running the models

In order to train the system

cd <MODEL_NAME>
python -m wikidata_query.train

Similarly for testing

cd <MODEL_NAME>
python -m wikidata_query.test

Citing the paper


@ARTICLE{2018arXiv181009164C,
   author = {{Cetoli}, A. and {Akbari}, M. and {Bragaglia}, S. and {O'Harney}, A.~D. and 
	{Sloan}, M.},
    title = "{Named Entity Disambiguation using Deep Learning on Graphs}",
  journal = {ArXiv e-prints},
archivePrefix = "arXiv",
   eprint = {1810.09164},
 primaryClass = "cs.CL",
 keywords = {Computer Science - Computation and Language},
     year = 2018,
    month = oct,
   adsurl = {http://adsabs.harvard.edu/abs/2018arXiv181009164C},
  adsnote = {Provided by the SAO/NASA Astrophysics Data System}
}

ned-graphs's People

Stargazers

Watchers

Forkers

codes-kzhan jdc08161063 sidney1994 tyut2018 currylym novellll chaitaliprabhu fractalego liqijialqj greitzmann samprintz rashmi9900 iurshina bxre7at7sf-forks

ned-graphs's Issues

Clarification on inputs to test the model

Error while parsing the rdf files to extract triplets.

Cannot reproduce results as mentioned in paper

I am unable to reproduce the results as mentioned in your paper on the same dataset, code and embeddings.
I have followed the exact steps mentioned in the readme and run the model "wikidata_entity_linking_with_attentive_rnn_triplets".
I haven't made any changes in the code and am using glove embeddings as mentioned, still the results do not match.
Kindly look into the issue.

Error

Have response like this and not have trained file ( Run in colab ) what happen ?
thank you

Seek information regarding the input parameters for the test setting

We seek information from your side to understand what are the input parameters in the test phase of the model:

Let's assume we have an input sentence "Michelle is walking on the street". Michelle is the surface form which has Its correct Wikidata ID is Q13133 (Michelle Obama)
while you test the model on the test set, what is the input? Do you just pass the sentence "Michelle is walking on the street" and directly predict its correct Wikidata Q id? Or you pass the sentence "Michelle is walking on the street" and also assume that Michelle as the recognized entity and pass sentence plus "Michelle" surface form as input and predicts the correct Wikidata ID?
We are a bit confused about the input values for the test cases. For the training, it is quite clear.

Any data for the KG triplets available?

I was wondering whether any information about the KG triplets was available anywhere?
I noticed that you're a using GCNs, thus I assume you should have constructed the graph associated to the entity linking task somewhere, am I right?
Indeed, I believe all the scipts implementing the SPARQL queries can be found in the corresponding wikidata_query directories, how am I supposed to use those to retrieve the triplets and build the Knowledge Graph?

Have probrem ..

i use library this

tensorflow 1.13.2
transformers==4.10.2
tensorflow==1.13.2
spacy==2.2.4
gensim==3.8.1

run in colab
when run train.py have error response like this , What happen ?

.. Thank you

error

thank you fore response

Have problem

I following
cd data
bunzip2 x*
cat x* > wikidata_items.csv
cd ..
after have wikidata_items.csv i try run train.py but have error below

what happen ?
thank you

Recommend Projects

React

A declarative, efficient, and flexible JavaScript library for building user interfaces.
Vue.js

🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
Typescript

TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
TensorFlow

An Open Source Machine Learning Framework for Everyone
Django

The Web framework for perfectionists with deadlines.
Laravel

A PHP framework for web artisans
D3

Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

javascript

JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
web

Some thing interesting about web. New door for the world.
server

A server is a program made to process requests and deliver data to clients.
Machine learning

Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Visualization

Some thing interesting about visualization, use data art
Game

Some thing interesting about game, make everyone happy.

Recommend Org

Facebook

We are working to build community through open source technology. NB: members must have two-factor auth.
Microsoft

Open source projects and samples from Microsoft.
Google

Google ❤️ Open Source for everyone.
Alibaba

Alibaba Open Source for everyone
D3

Data-Driven Documents codes.
Tencent

China tencent open source team.