mesnilgr / nbsvm Goto Github PK

Naive Bayes SVM for Sentiment Analysis

Python 68.87% Shell 31.13%

nbsvm's Introduction

Note: I don't provide personal support for custom changes in the code. Only for the release. For people just starting, I recommend Treehouse for online-learning.

Naive Bayes SVM (NB-SVM)

This code reproduces performance of the NB-SVM on the IMDB reviews from the paper:

Sida Wang and Christopher D. Manning: Baselines and Bigrams: Simple, Good Sentiment and Topic Classification; ACL 2012. http://nlp.stanford.edu/pubs/sidaw12_simple_sentiment.pdf

They obtain 91.22% while this code obtains 91.55% with bigrams and 91.82% with trigrams. Little improvements (+0.33% with bigrams and +0.6% with unigrams) versus the paper.

To reproduce the results:

git clone [email protected]:mesnilgr/nbsvm.git
cd nbsvm; chmod +x oh_my_go.sh
./oh_my_go.sh

End to end (downloading the data, tokenizing, training the models), this will take 68 mins. Note that most of the time is spent dowloading and tokenizing. Once the data has been downloaded and tokenized, training an NB-SVM only takes ~2 mins for uni+bigrams and <5 mins for uni+bi+trigrams.

Naive Bayes SVM by Grégoire Mesnil is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
Based on a work at https://github.com/mesnilgr/nbsvm.

nbsvm's People

Contributors

Stargazers

Watchers

Forkers

xmzhao zhoujialinmumu ty01csbaidu xuanqing94 bachlog dpressel milesqli hfoffani mlwave huangtao030 zbxzc35 zuiwufenghua le02146 milestonesvn weiyudang nkhuyu snazz2001 technologiclee jobsformek jeanvc stevenlol fbkarsdorp zach2015 ml-ai-nlp-ir nlplab908 imclab dqgv587 fangzheng354 dave1453629500 wdw110 jeep123 minhpqn broshanfekr deepakkr249 thanhlct susuisui cyber-neuron anranwang mjewkes aniketgurav mazhengm2 solertis zengxia alxsoares timwee mjahanshahi michaelcapizzi kastnerkyle jizhihang wanesta sandy4321 gupineee tarogege mayccc veerendrab ujjwalmishra zackseyun pmil0924 usingmeee soapwang afcarl zouzias votamvan b2220333 kaji46 rakshpar vinodkandula liushixing zpli0320 auserj yanghaha11514 liyunchao mazib dengyuememeda tashengjinsheng wuhaodemo jianbo-lab ykhelloword xuzhezhaozhao tdhworkspace shikhasahu zhenjason yangyang233 kyoungrok0517 leiurus17 harshitharavindra29 fariedrahmat datatalking aisha-abdool pengchen233 susululu idafatriniputri arklu

nbsvm's Issues

support for multi class task

we can see from your code and paper that your method can do a verry good job on the binary class text classification tasks.But if we need to solve a multi class issue,how can we promote your algorithom
and the code? Thank you~

Install instruction yield ../nbsvm: No such file or directory

When I follow the instructions to this otherwise helpful Naive Bayes python script it gives me an error.

owner at Owners-iMac in ~/sbox/test/nbsvm on master
$ ./oh_my_go.sh
mkdir: nbsvm_run: File exists
dyld: Library not loaded: /usr/local/opt/openssl/lib/libssl.1.0.0.dylib
  Referenced from: /usr/local/bin/wget
  Reason: image not found
./oh_my_go.sh: line 11: 64166 Abort trap: 6           wget http://ai.stanford.edu/~amaas/data/sentiment/aclImdb_v1.tar.gz
tar: Error opening archive: Failed to open 'aclImdb_v1.tar.gz'
rm: aclImdb_v1.tar.gz: No such file or directory
ls: aclImdb/train/pos: No such file or directory
./oh_my_go.sh: line 3: temp: No such file or directory
mv: rename temp-norm to aclImdb/train/pos/norm.txt: No such file or directory
rm: temp: No such file or directory
ls: aclImdb/train/neg: No such file or directory
./oh_my_go.sh: line 3: temp: No such file or directory
mv: rename temp-norm to aclImdb/train/neg/norm.txt: No such file or directory
rm: temp: No such file or directory
ls: aclImdb/test/pos: No such file or directory
./oh_my_go.sh: line 3: temp: No such file or directory
mv: rename temp-norm to aclImdb/test/pos/norm.txt: No such file or directory
rm: temp: No such file or directory
ls: aclImdb/test/neg: No such file or directory
./oh_my_go.sh: line 3: temp: No such file or directory
mv: rename temp-norm to aclImdb/test/neg/norm.txt: No such file or directory
rm: temp: No such file or directory
mkdir: data: File exists
mv: rename aclImdb/train/pos/norm.txt to data/train-pos.txt: No such file or directory
mv: rename aclImdb/train/neg/norm.txt to data/train-neg.txt: No such file or directory
mv: rename aclImdb/test/pos/norm.txt to data/test-pos.txt: No such file or directory
mv: rename aclImdb/test/neg/norm.txt to data/test-neg.txt: No such file or directory
rm: aclImdb: No such file or directory
dyld: Library not loaded: /usr/local/opt/openssl/lib/libssl.1.0.0.dylib
  Referenced from: /usr/local/bin/wget
  Reason: image not found
./oh_my_go.sh: line 29: 64195 Abort trap: 6           wget https://www.csie.ntu.edu.tw/~cjlin/liblinear/oldfiles/liblinear-1.96.zip
unzip:  cannot find or open liblinear-1.96.zip, liblinear-1.96.zip.zip or liblinear-1.96.zip.ZIP.
rm: liblinear-1.96.zip: No such file or directory
./oh_my_go.sh: line 32: cd: liblinear-1.96: No such file or directory
make: *** No targets specified and no makefile found.  Stop.
BI-GRAM
python: can't open file '../nbsvm/nbsvm.py': [Errno 2] No such file or directory
TRI-GRAM
python: can't open file '../nbsvm/nbsvm.py': [Errno 2] No such file or directory
./oh_my_go.sh: line 40: cd: ../nbsvm: No such file or directory
(base)

what external code you use?

os.system(trainsvm + " -s 0 train-nbsvm.txt model.logreg")

is it libsvm, if yest how to install on windows

mesnilgr / nbsvm Goto Github PK

nbsvm's Introduction

Naive Bayes SVM (NB-SVM)

nbsvm's People

Contributors

Stargazers

Watchers

Forkers

nbsvm's Issues

support for multi class task

Install instruction yield ../nbsvm: No such file or directory

what external code you use?

Recommend Projects

React

Vue.js

Typescript

TensorFlow

Django

Laravel

D3

Recommend Topics

javascript

web

server

Machine learning

Visualization

Game

Recommend Org

Facebook

Microsoft

Google

Alibaba

D3

Tencent