Hi Phong Tran N., I noticed your profile and would like to offer you my project. We can discuss any details over chat.
I am working on this paper: http://shebuti.com/wp-content/uploads/2016/06/15-kdd-collectiveopinionspam.pdf for which the Matlab and Python 2 code is provided here:
https://www.dropbox.com/sh/iqcuj0363zcj3go/AAAvbZVR_PSNyJX8AXUXpBqea?dl=0. In the code there is a file called featureExtraction.m in which the features described in paper are extracted from data file. This file has references to various functions (again in the Dropbox) to get the features.
I want all of this code related to feature extraction ( and related functions) to be rewritten in Python 3. Additionally I have a couple of more features that require additional Python 3 code to be written ( these are not in Matlab code). The complete list of features I need is in the Excel file attached (most of which are coming from the paper so matlab code is present).
The additional features requires knowledge of Opinion Mining - Sentiment Analysis ( Especially Aspect Based Sentiment Analysis).
The data file on which this Python 3 code should be run is present here: https://drive.google.com/open?id=0B8JIKvhJUvRdfk8yS1E4T0lXUm1uOGtJUmN2cExMTXRmVUpsSGE2OHRzNkdUT0RyMzA4WDA (reviewContent and metadata)
As a deliverable I will require Python 3 code to extract features from the data file and a Text file which has the features extracted ( 1 review per line and columns corresponds to the features and whether the review is fake or not).
The developer that will work on this project should have extensive knowledge of Python libraries for text mining and sentiment analysis and working knowledge of Matlab.