MétaCan
Menu
Back to cohort
Record W7007999199

ARTIFICIAL NEURAL NETWORK (ANN) APPLIED TO THE SRC DIAGNOSTIC TOOL

2002· dissertation· en· W7007999199 on OpenAlexaffabout

Bibliographic record

VenueUniversity Library (University of Saskatchewan) · 2002
Typedissertation
Languageen
FieldComputer Science
TopicNeural Networks and Applications
Canadian institutionsUniversity of Saskatchewan
Fundersnot available
KeywordsArtificial neural networkSet (abstract data type)Process (computing)Data setComponent (thermodynamics)Range (aeronautics)Time delay neural network
DOInot available

Abstract

fetched live from OpenAlex

This study determined the feasibility of applying artificial neural networks to a business diagnostic tool. Artificial neural networks (ANN) are a component of intelligent systems technologies and have been applied to wide range of disciplines ranging from engineering to business. In this study, an artificial neural network was to be used to model the analysis process that was manually performed after a set of diagnostic surveys were completed. The specific diagnostic survey, called the Diagnostic Tool, was developed by the Saskatchewan Research Council and the College of Commerce. The Diagnostic Tool was used to gather information from small and medium sized enterprises and, after analysis, to provide the companies with feedback on how they could improve operations. The manual analysis process was tedious and suffered from inconsistencies thus creating an interest in finding alternate methods of analysis. During the course of this study, three artificial neural network models were created. These models differed primarily by the number of inputs and type of input data. The first model used all the inputs from a section (Large model) and the second (Trimmed model) used a reduced set of inputs, determined from contribution weightings. The third model (Optimized model) used a reduced set of inputs but information from all inputs were combined into the reduced input data set. Because of insufficient actual company data, a simulation data set of 300 companies was created. Manual analysis on the 300 simulation data sets was performed and this information was provided to the artificial neural network models. For comparison purposes, a multiple regression model was created with each ANN to determine, in general, how each methodology compared. Two sections of the Diagnostic Tool were used in the study, the Human Resources and Manufacturing Management sections. In addition, a survey of business literature was conducted for the purpose of determining what practices benefit a small and medium sized enterprise. The information derived from this study was to be used in updating the Diagnostic Tool. Of the three ANN models, the Optimized model performed best. This model showed the lowest mean error for both the Human Resources and Manufacturing Management sections. The regression equations developed along with each ANN model showed similar performance to the ANN but in almost all cases the ANN models outperformed the regression equations. The ANN models were easier to implement and modify compared to the regression equations. It was observed that the results for Manufacturing Management were better (by almost 7%) than for Human Resources. This was attributed to the fact that because the Manufacturing Management section was rated by the analyst after the Human Resources section, the analyst, improved his/her rating consistency. Out of the sample mix of companies, those with average performance showed the highest standard deviation, compared to companies with very good or poor performance. This was due to the wider range of values the analyst could give to an average company.

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 imitation

Not 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.

metaresearch head score (Codex)0.004
metaresearch head score (Gemma)0.019
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Simulation or modeling · Consensus signal: Simulation or modeling
GenreCandidate signal: Empirical · Consensus signal: none
Teacher disagreement score0.012
Threshold uncertainty score0.025

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.0040.019
Meta-epidemiology (narrow)0.0010.000
Meta-epidemiology (broad)0.0010.001
Bibliometrics0.0020.002
Science and technology studies0.0000.000
Scholarly communication0.0010.001
Open science0.0010.001
Research integrity0.0010.001
Insufficient payload (model declined to judge)0.0030.001

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.

Opus teacher head0.008
GPT teacher head0.167
Teacher spread0.158 · 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

Classification

machine, unvalidated

Machine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.

The models applied no category: nothing in the taxonomy fit this work.
Study designSimulation or modeling
Domainnot available
GenreEmpirical

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".

Quick stats

Citations0
Published2002
Admission routes2
Has abstractyes

Explore more

Same venueUniversity Library (University of Saskatchewan)Same topicNeural Networks and ApplicationsFrench-language works237,207