Twitter Texts’ Quality Classification using Data Mining and Neural Networks
Bibliographic record
Abstract
Purpose: This is an attempt to classify the level of noise in twitter texts which is part of social media data analytics problem.Estimations in recent machine learning & data feeding algorithms researches' assumptions consider high data quality in social media texts, while they actually lack data accuracy, completeness, and overall quality which leads to the principle of "Garbage In Garbage Out" resulting in bizarre statistical findings.The aim of this project is to predict and classify Twitter data noise levels using a labelled dataset.Methodology: After data cleaning, a clustering technique was used to find the major dimensions in the data imported, and a dimension reduction algorithm was ran using PCA Weighting and the Wight Guided Feature Selection algorithms.They resulted into 6 most significant features which were used in the implementation.An artificial neural network model was trained to predict the Tweets' quality classes using R and RStudio.The ANN used is Neural Network (NN) and Naïve Bayes (NB) for the purpose of predicting the Twitter text quality.There will be a comparison between the 2 ANN used in terms of accuracy and precision.Findings: Three different aspects of text mining were discovered in twitter data.(1) Neural network gives surprisingly good result as compared to Naive Bayes algorithm, (2) With only 3 hidden layers, a network was created which can predict good or bad class, (3) Preprocessing of the data and implementing predictive algorithms take huge data and very high computational complexity and time.Research results show that Neural Network performs well even without Dropout layer and convolutional layers.The accuracy of the Neural Network is 99%.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.009 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.005 | 0.002 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".