MétaCan
Menu
Back to cohort
Record W4412714478 · doi:10.59543/ijmscs.v3i.15097

Comparative Analysis of AI Models for Effort Estimation in Western and Regional Environments

2025· article· en· W4412714478 on OpenAlex

Why this work is in the frame

A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.

aboutThe title or abstract carries a Canadian signal from the geographic lexicon.
no affNo Canadian affiliation: this work is invisible to an affiliation-only frame.
No Canadian affiliation. An affiliation-only frame, the usual design, would never have seen this work. It is one of the works that make the case for inverting the frame.

Bibliographic record

VenueInternational Journal of Mathematics Statistics and Computer Science · 2025
Typearticle
Languageen
FieldComputer Science
TopicSoftware Engineering Research
Canadian institutionsnot available
Fundersnot available
KeywordsMean squared errorRandom forestWorkflowEstimationArtificial neural networkWork (physics)SoftwareComputer scienceDocumentationOperations researchArtificial intelligenceStatisticsEngineeringMathematicsEconomicsManagement

Abstract

fetched live from OpenAlex

Artificial Intelligence rapidly alters business operations and workflow management strategies these days in various corporate sectors. AI can be effective in calculating the effort required for a software. Figuring that how much time work and resources some project will probably take requires a fairly decent amount of effort and is also one of the most crucial activities of software project management. When an estimate is met the confidence of companies increase and funds can be allocated nicely, ultimately helping in finishing projects quickly. This work evaluates different AI models for estimating software effort accurately in two distinct areas. Western environment encompasses nations such as US and UK, and mostly all developed places including Canada alongside other similar countries. These countries typically possess sophisticated technology and proficient labor with meticulous documentation practices. Regional environment encompasses areas namely South Asia and Africa alongside Middle East which undergo development challenges. These areas often face many problems such as weak digital infrastructure in various sectors and somewhat disorganized data sets which are not very helpful for estimation. Various AI models were tested including Linear Regression, Neural Networks random forest etc. in different areas to determine which ones worked nicely. Three measures were used namely Mean Absolute Error (MAE) Root Mean Squared Error (RMSE) and R² Score to assess AI model efficiency. Better accuracy stems from lower MAE and RMSE values while higher R² scores signify deeper understanding of data patterns. Neural Networks operate more effectively in Western regions owing largely to relatively cleaner data and markedly greater regularity. Random Forests and Decision Trees perform markedly better in regional areas plagued by messy data because they handle such info pretty well. Companies ought to select an AI model suited pretty well to their specific local conditions and the kind of data they possess. Finally in both the environments, that means deploying both the datasets for the environments, it was the hybrid technique that performed the best for predicting the effort of software. The hybrid model used for prediction give the lowest MAE of 0.22 and |RMSE of 0.38 with R2 of 0.9 for the Western regions. Similary even for the regional areas give the lowest MAE and RMSE of 0.4 and 0.55 respectively and R2 of 0.79.

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.

Full frame distilled prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.

metaresearch head score (Codex)0.001
metaresearch head score (Gemma)0.000
Version: codex-gemma-dda1882f352aValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Simulation or modeling · Consensus signal: none
GenreCandidate signal: Methods · Consensus signal: none
Teacher disagreement score0.609
Threshold uncertainty score0.256

Codex and Gemma teacher scores by category

CategoryCodexGemma
Metaresearch0.0010.000
Meta-epidemiology (narrow)0.0000.000
Meta-epidemiology (broad)0.0000.000
Bibliometrics0.0010.000
Science and technology studies0.0000.000
Scholarly communication0.0000.000
Open science0.0010.000
Research integrity0.0000.000
Insufficient payload (model declined to judge)0.0000.000

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.

Opus teacher head0.028
GPT teacher head0.343
Teacher spread0.314 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it