Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Blog Data Showdown: Machine Learning vs Neuro-Symbolic Models for Gender Classification

Natnael Tilahun Sinshaw, Mengmei He, Tadesse K. Bahiru, Sudhir Kumar Mohapatra

TL;DR

This paper evaluates gender classification from blogs by comparing traditional ML, deep learning, and neuro-symbolic (NeSy) models using the Blog Author Dataset. It explores multiple text representations (TF-IDF, USE, RoBERTa) and feature-selection schemes (Chi-Square, MI, PCA), including handcrafted cues, and introduces a Logic Tensor Network–driven NeSy architecture paired with an MLP. Key findings show SVM with PCA+Chi-Square attaining the strongest baseline accuracy (~0.78), while NeSy achieves competitive performance (≈0.75) and higher ROC (≈0.81) but does not surpass the SVM baseline yet. The results suggest NeSy is a promising approach for small datasets and combine symbolic reasoning with neural models, with future work on expanding embeddings, datasets, and hyperparameters.

Abstract

Text classification problems, such as gender classification from a blog, have been a well-matured research area that has been well studied using machine learning algorithms. It has several application domains in market analysis, customer recommendation, and recommendation systems. This study presents a comparative analysis of the widely used machine learning algorithms, namely Support Vector Machines (SVM), Naive Bayes (NB), Logistic Regression (LR), AdaBoost, XGBoost, and an SVM variant (SVM_R) with neuro-symbolic AI (NeSy). The paper also explores the effect of text representations such as TF-IDF, the Universal Sentence Encoder (USE), and RoBERTa. Additionally, various feature extraction techniques, including Chi-Square, Mutual Information, and Principal Component Analysis, are explored. Building on these, we introduce a comparative analysis of the machine learning and deep learning approaches in comparison to the NeSy. The experimental results show that the use of the NeSy approach matched strong MLP results despite a limited dataset. Future work on this research will expand the knowledge base, the scope of embedding types, and the hyperparameter configuration to further study the effectiveness of the NeSy approach.

Blog Data Showdown: Machine Learning vs Neuro-Symbolic Models for Gender Classification

TL;DR

This paper evaluates gender classification from blogs by comparing traditional ML, deep learning, and neuro-symbolic (NeSy) models using the Blog Author Dataset. It explores multiple text representations (TF-IDF, USE, RoBERTa) and feature-selection schemes (Chi-Square, MI, PCA), including handcrafted cues, and introduces a Logic Tensor Network–driven NeSy architecture paired with an MLP. Key findings show SVM with PCA+Chi-Square attaining the strongest baseline accuracy (~0.78), while NeSy achieves competitive performance (≈0.75) and higher ROC (≈0.81) but does not surpass the SVM baseline yet. The results suggest NeSy is a promising approach for small datasets and combine symbolic reasoning with neural models, with future work on expanding embeddings, datasets, and hyperparameters.

Abstract

Text classification problems, such as gender classification from a blog, have been a well-matured research area that has been well studied using machine learning algorithms. It has several application domains in market analysis, customer recommendation, and recommendation systems. This study presents a comparative analysis of the widely used machine learning algorithms, namely Support Vector Machines (SVM), Naive Bayes (NB), Logistic Regression (LR), AdaBoost, XGBoost, and an SVM variant (SVM_R) with neuro-symbolic AI (NeSy). The paper also explores the effect of text representations such as TF-IDF, the Universal Sentence Encoder (USE), and RoBERTa. Additionally, various feature extraction techniques, including Chi-Square, Mutual Information, and Principal Component Analysis, are explored. Building on these, we introduce a comparative analysis of the machine learning and deep learning approaches in comparison to the NeSy. The experimental results show that the use of the NeSy approach matched strong MLP results despite a limited dataset. Future work on this research will expand the knowledge base, the scope of embedding types, and the hyperparameter configuration to further study the effectiveness of the NeSy approach.

Paper Structure

This paper contains 17 sections, 7 equations, 6 figures, 3 tables.

Table of Contents

  1. Introduction
  2. Related Works
  3. Methodology
  4. Dataset Sources
  5. Data Preprocessing
  6. Feature Engineering
  7. Word Embeddings
  8. Algorithm Selection
  9. ML Baseline
  10. The NeSy Architecture
  11. MLP Components
  12. LTN Formulation
  13. Experiments
  14. Results
  15. Discussion
  16. ...and 2 more sections

Figures (6)

  • Figure 1: Feed-forward network without the third hidden layer
  • Figure 2: Overall Architecture
  • Figure 3: Training and Validation Accuracy Curve
  • Figure 4: Training and Validation Accuracy Curve
  • Figure 5: Training & Validation Accuracy/LTN Saturation
  • ...and 1 more figures