Classification of Poverty Condition Using Natural Language Processing

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216208%3A11320%2F22%3A88HHHD7W" target="_blank" >RIV/00216208:11320/22:88HHHD7W - isvavai.cz</a>

Výsledek na webu

<a href="https://doi.org/10.1007/s11205-022-02883-z" target="_blank" >https://doi.org/10.1007/s11205-022-02883-z</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1007/s11205-022-02883-z" target="_blank" >10.1007/s11205-022-02883-z</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Classification of Poverty Condition Using Natural Language Processing

Popis výsledku v původním jazyce

This work introduces a methodology to classify between poor and extremely poor people through Natural Language Processing. The approach serves as a baseline to understand and classify poverty through the people’s discourses using machine learning algorithms. Based on classical and modern word vector representations we propose two strategies for document level representations: (1) document-level features based on the concatenation of descriptive statistics and (2) Gaussian mixture models. Three classification methods are systematically evaluated: Support Vector Machines, Random Forest, and Extreme Gradient Boosting. The fourth best experiments yielded around 55% of accuracy, while the embeddings based on GloVe word vectors yielded a sensitivity of 79.6% which could be of great interest for the public policy makers to accurately find people who need to be prioritized in social programs.

Název v anglickém jazyce

Classification of Poverty Condition Using Natural Language Processing

Popis výsledku anglicky

This work introduces a methodology to classify between poor and extremely poor people through Natural Language Processing. The approach serves as a baseline to understand and classify poverty through the people’s discourses using machine learning algorithms. Based on classical and modern word vector representations we propose two strategies for document level representations: (1) document-level features based on the concatenation of descriptive statistics and (2) Gaussian mixture models. Three classification methods are systematically evaluated: Support Vector Machines, Random Forest, and Extreme Gradient Boosting. The fourth best experiments yielded around 55% of accuracy, while the embeddings based on GloVe word vectors yielded a sensitivity of 79.6% which could be of great interest for the public policy makers to accurately find people who need to be prioritized in social programs.

Druh

J_imp - Článek v periodiku v databázi Web of Science

CEP obor

—

OECD FORD obor

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt

—

Návaznosti

—

Rok uplatnění

2022

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název periodika

Social Indicators Research [online]

ISSN

1573-0921

e-ISSN

1573-0921

Svazek periodika

162

Číslo periodika v rámci svazku

3

Stát vydavatele periodika

JP - Japonsko

Počet stran výsledku

23

Strana od-do

1413-1435

Kód UT WoS článku

000752794700001

EID výsledku v databázi Scopus

2-s2.0-85124366997