A Deep Neural Network Optimized by a Genetic Algorithm to Improve Arabic Sentiment Classification
Bibliographic record
Abstract
Deep learning has improved the state-of-the-art in sentiment analysis for various languages, including Arabic.One aspect that can affect the performance of deep learning-based sentiment classification is the optimization method used for training the neural network.The conventional optimization method is carried out by a backpropagation (BP) algorithm that relies on gradient descent to find the minimum of a cost function.However, BP has the tendency to converge into local minima instead of global minima since neural networks generate complex error surfaces for even simple problems.In this study, for the purpose of improving the Arabic sentiment classification, we propose to use a genetic algorithm (GA) to train a deep neural network (DNN).GA is a meta-heuristic optimization algorithm inspired by the theory of natural evolution.The algorithm is expected to improve the classifier's performance due to its capability to reach optimal or near-optimal solutions.The proposed method uses Arabic sentiment lexicons to extract various features considering different aspects for text representation.The effectiveness of the proposed method is evaluated by analyzing its performance, versus a DNN trained with BP algorithm.The experimental results show that the proposed method can present better F1-measure of 90.7% for Arabic sentiment classification than traditional BP-based DNN.
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 | 0.001 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
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".