Using feedforward neural networks and forward selection of input variables for an ergonomics data classification problem

Abstract

A method was developed to accurately predict the risk of injuries in industrial jobs based on datasets not meeting the assumptions of parametric statistical tools, or being incomplete. Previous research used a backward-elimination process for feedforward neural network (FNN) input variable selection. Simulated annealing (SA) was used as a local search method in conjunction with a conjugate-gradient algorithm to develop an FNN. This article presents an incremental step in the use of FNNs for ergonomics analyses, specifically the use of forward selection of input variables. Advantages to this approach include enhancing the effectiveness of the use of neural networks when observations are missing from ergonomics datasets, and preventing overspecification or overfitting of an FNN to training data. Classification performance across two methods involving the use of SA combined with either forward selection or backward elimination of input variables was comparable for complete datasets, and the forward-selection approach produced results superior to previously used methods of FNN development, including the error back-propagation algorithm, when dealing with incomplete data.

Predicting the Risk of Low Back Disorders due to Manual Handling Tasks

“…Finally, the receiver operating characteristics (ROC) charts and the areas under them are used to determine the global predictive power of the created models at different (from standard 0.5) probability cut-off points from within the range [0,1]. Though the obtained classification accuracy rates for our best models are better than those reported in National Institute for Occupational Health and Safety (NIOHS) Guides [22], [23] and two of our previous studies [33], [34], they are generally less optimistic than those reported in several of the previous studies [1], [2], [6], [7], [8]; this work offers a more systematic and reliable approach for building and testing the performance of the classifiers to discriminate between high risk and low risk manual lifting tasks that cause LBDs.…”

“…This was actually the same NN-based model that was reported in their previous study [7]. Chen [8] also generally discussed the ROC charts for classification of LBDs in the context of a cost function and committing the type I and type II errors. The authors, however, neither created nor analyzed ROC charts generated by their models.…”

“…Their system showed a noticeable improvement, especially in classifying high risk jobs (83.9%). In their next study, Chen [8] performed more extensive computer simulation testing the performance of LR, multiple discriminant analysis (MDA), and NN. They used the original full data set with five input variables as well as the attribute reduction methods.…”

“…There have been a number of intensive research efforts [10], [11], [13], [14], [15], [18], [19], [26], [27], [28] devoted to understanding the nature and phenomena of LBDs as well as building classification models that could effectively distinguish between high risk and low risk MMH tasks that contribute to LBDs [1], [2], [6], [7], [8], [33], [34]. Despite more than two dozen years of studies, however, classifiers which could very effectively discriminate between low risk and high risk MMH that cause LBDs have not been created yet.…”

A Neural Network-Based System for Prediction of Computer User Comfort

“…(e.g., Karwowski, Jarvinen, and Zurada 1992;Arcand 1994;Jung and Park 1994;Lim, Fok, and Tan 1996;Harden, Crumpton, and Killough 1996;Zurada, Karwowski, and Marras 1997;Carnahan and Redfern 1998;Lin and Hwang 1999;Chen, Kaber, and Dempsey 2000;Chung, Lee, Inseok, Dohyung, and Sang 2002;Kiryu, Shibai, Hayashi, and Tanaka 2002;Kaya et al 2003;Lee, Karwowski, Marras, and Rodrick 2003;Chen, Kaber, and Dempsey 2004;Kolich, Seal, and Taboun 2004). Chen et al (2004) used neural networks for data classification problems. They developed a method to predict the risk of inquiries in industrial jobs.…”