https://github.com/alrobles/mammals_virus_text_class

Science Score: 10.0%

This score indicates how likely this project is to be science-related based on various indicators:

○
CITATION.cff file
○
codemeta.json file
○
.zenodo.json file
○
DOI references
✓
Academic publication links
Links to: pubmed.ncbi, ncbi.nlm.nih.gov
○
Academic email domains
○
Institutional organization owner
○
JOSS paper metadata
○
Scientific vocabulary similarity
Low similarity (5.3%) to scientific vocabulary

Last synced: 10 months ago · JSON representation

Repository

Basic Info

Host: GitHub
Owner: alrobles
Language: R
Default Branch: main
Size: 1.91 MB

Statistics

Stars: 0
Watchers: 1
Forks: 0
Open Issues: 0
Releases: 0

Created about 4 years ago · Last pushed almost 3 years ago

https://github.com/alrobles/mammals_virus_text_class/blob/main/

Mammal parasite paper recommender
================

## Methodology

This is a brief summary of the methodology we use

-   From global parasite databse () we extract all the papers  
-   Because we only have the name title of the paper we extract the
    title and search in Cross Reference ()
    the doi and the abstract if is avaliable.
-   Also we search with the DOI and title in PubMed
    ()
-   We finally build a table with title, DOI, abstract and year.
-   We label this information as gmpd
-   We generate a 1000 random abstracts sample in order to has a second
    class we call unknown
-   We tokenize each abstract and create a vocabulary for each word in
    abstract
-   We also take in accout the bigrams and remove the common stoping
    words in english.
-   We keep with a vocabulary of terms with a minimum term count of 20
    overall the papers
-   We vectorize or vocabulary and create a document term matrix wich
    contains
-   With the DTM we train a penalized logistic regression model. We
    model the two clases (gmpd and uknown) and interpret the probability
    as a measure of how an abstracts has information close to GMPD or
    not.
-   We build a shiny app that search an arbitrary string in PubMed and
    evaluates the abstracts of each paper founded acoording to our
    linear model

## App deploy

The app is accesible in

Owner

Name: Angel Luis Robles Fernández
Login: alrobles
Kind: user
Location: Xalapa Mexico
Company: Vida Analytics

ecosyste.ms

Data

Tools

Indexes

Applications

Experiments

Open Source Science