A hybrid model combining neural networks and decision tree for comprehension detection.
read more
Citations
The politics of deceptive borders: ‘biomarkers of deceit’ and the case of iBorderCtrl
The association between 5-HTTLPR and spontaneous facial mimicry: An investigation using the Facial Action Coding System (FACS)
Data mining for assessing the credit risk of local government units in Croatia
Reconciling Adapted Psychological Profiling with the New European Data Protection Legislation
The politics of deceptive borders: 'biomarkers of deceit' and the case of iBorderCtrl
References
Comprehension: A Paradigm for Cognition
Nonverbal communication in human interaction
The Career of Metaphor.
Bosphorus Database for 3D Face Analysis
Trees vs Neurons: Comparison between random forest and ANN for high-resolution prediction of building energy consumption
Related Papers (5)
Frequently Asked Questions (16)
Q2. What future works have the authors mentioned in the paper "A hybrid model combining neural networks and decision tree for comprehension detection" ?
CONCLUSIONS AND FUTURE WORK The authors propose that future work should explore several options to simplify the decision trees and their representation. Second, investigation of the potential to reduce the number of input channels – through empirical experiment, by identifying the potentially lowest contributing channels through calculating information content and by grouping channels.
Q3. What contributions have the authors mentioned in the paper "A hybrid model combining neural networks and decision tree for comprehension detection" ?
This paper investigates the use of a hybrid model comprising multiple artificial neural networks with a final C4. 5 decision tree classifier to investigate the potential of explaining the classification decision through production rules.
Q4. What have the authors stated for future works in "A hybrid model combining neural networks and decision tree for comprehension detection" ?
CONCLUSIONS AND FUTURE WORK The authors propose that future work should explore several options to simplify the decision trees and their representation. Second, investigation of the potential to reduce the number of input channels – through empirical experiment, by identifying the potentially lowest contributing channels through calculating information content and by grouping channels.
Q5. What was the purpose of the study?
For each study, FATHOM’s object locators and pattern detectors were used to extract and collate the non-verbal vector-based dataset for the purpose of training the final BPANN classifier.
Q6. What is the definition of non-verbal behaviour?
Non-verbal behaviour comprises all of the signals or cues, which human beings use to communicate, including visual, audio, tactile and chemical components [16, 17].
Q7. How do the authors reduce the rule sets extracted from the more efficient trees?
by using fuzzy rule extraction or random forest techniques to reduce the rule sets extracted from the more efficient trees to a more tractable size.
Q8. How many participants were selected to participate in the study?
Forty participants were selected to participate in the study, from academic and technical staff at the Manchester Metropolitan University (MMU) in the UK.
Q9. How is the input to FATHOM offline?
Input to FATHOM is currently offline through recorded videos, which are streamed into FATHOM where a series of BPANN facial object locators, identify the location in a video frame of key visual features such as the eyes.
Q10. What was the initial range used for the pruning experiments?
The initial ranges used for the experiments were, for CI: 0.25,0.2, 0.15, 0.1, 0.05, and for MNO: 2, 5, 10, 15, 20.V. RESULTSTable
Q11. How many vectors were included in the study?
Cross-validation: 10-foldsFor study 1, eighty randomly selected participant videos ( from the 292 obtained in the study) comprised the HIV Informed Consent dataset containing 71,787 vectors with 63.5% comprehension and 36.5% non-comprehension.
Q12. What was the training method used to train the BPANN classifier?
The following training parameters (determined from previous exploratory cross-validation sessions) were used to train the single hidden layer neural network in the Fathom training application: Topology: 40:20:1 Accept value: 1.0 (output >= 0.0 equals comprehension AND output <0.0 equals non-comprehension) Maximum epochs: 10,000 Checking epochs: 250, i.e. at every 250th epoch the total Classification accuracy (CA) was checked and if there wasno improvement training was terminated.
Q13. What is the way to clean up the data?
pre-preprocessing the data to cleanse it, particularly removing outliers, noise and conflicting records - all of which might be better handled by the BPANN than DTs.
Q14. How many participants were invited to participate in the study?
Each participant was invited to engage individually in a short learning task, which was comprised of watching a short video on Termites and then answering a small set of associated assessment questions whilst being video recorded.
Q15. What is the purpose of the paper?
The work presented in this paper investigates the consequences of replacing the BPANN comprehension classifier in the FATHOM system by a C4.5 decision tree [12], to answer questions about their relative performance and transparency.
Q16. What was the purpose of the experiment?
The experimental methodology was to take the pair ofdatasets outlined in Section III and use them to train andevaluate C4.5 decision trees to replace the final stage BPANNclassifier.