Leveraging Latent Dirichlet Allocation for Feature Extraction in User Comments: Enhancements to User-Centered Design in Indonesian Financial Technology
Bibliographic record
Abstract
The burgeoning Financial Technology (FinTech) sector in Indonesia, while witnessing a surge in user base, contends with limitations in system functionalities and service offerings.The extraction of latent features from user comments is imperative for the identification of these inadequacies, serving as a catalyst for innovation and delivering advantages to both consumers and developers of FinTech applications.This study employs Latent Dirichlet Allocation (LDA) algorithm, an intelligent probabilistic model, to discern and extract underlying topics within narratives found in user comments on FinTech platforms.In this approach, words within each topic are ranked according to their respective probabilities.Through the LDA algorithm, ten salient topics, each comprising ten keywords, have been identified.These topics coalesce into three broad categories: system improvements in applications, services that are out of sync with the system, and service satisfaction.The coherence of the topics has been quantitatively assessed, with an average score of 0.564, indicative of substantial coherence.Findings from the LDA model are integrated into the User-Centered Design (UCeD) framework, wherein the algorithm streamlines the UCeD's evaluative and abstraction processes, as well as the grouping of user necessities.This integration aids FinTech management teams in pinpointing pertinent user feedback, thereby facilitating the refinement of application development.
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 | 0.012 | 0.029 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.003 | 0.002 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.003 | 0.003 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 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".