An efficient Concealed Information Test: EEG feature extraction and ensemble classification for lie identification

The Concealed Information Test (CIT), also known as the Guilty Knowledge Test (GKT), is a method of deception detection by presenting the suspects with familiar stimuli that only the culprits would recognize. Because recognition memory is crucial to the success of reaction time-based CIT, in this study we investigated how pictorial and verbal crime-related items (i.e., Probes) of a visually experienced mock crime would perform against time delay and memory deterioration. Participants visually encoded the details of a theft mock crime from a first-person perspective and were randomly assigned to one of the Probe modalities (i.e., verbal vs. pictorial) group and one of the time delays (immediate vs. 2-week delay) group: immediate verbal, immediate pictorial, 2-week verbal, 2-week pictorial. We observed significant Probe-Irrelevant RT difference in all 4 conditions. When we evaluated the ability of each condition to differentiate between guilty and innocent by calculating the area under the curve (AUC) with Receiver Operating Characteristic analysis, we observed a significant effect of time delay, with higher AUCs in the immediate condition compared to the 2-week delay, and no difference between verbal and pictorial modalities. Together, our results suggest that for visually experienced events, CIT effects can be observed 1) both with pictorial or verbal Probes, 2) both immediately and after 2 weeks, and 3) importantly, although pictorial and verbal Probes perform equally well, a 2-week delay would cost CIT efficacy in AUC.

Chapter 21 - The Instrumental Detection of Deception: Polygraph

Electroencephalogram (EEG) signal is of- ten contaminated by electronic noise as well as move- ment artifacts. This paper presented an algorithm based on Canonical correlation analysis (CCA) to estimate multi- channel EEG data. Different from previous studies, in which CCA was mainly used to detect the invariant fea- tures specific to each brain state, in this paper, the canon- ical variates computed by CCA were used to reconstruct the multi-channel EEG data. Firstly, two data sets, EEG signals and the reference signals based on prior knowledge were constructed. Next, canonical variates were computed by projecting the two data sets onto basis vectors. Finally, a least squares solution was used to estimate the multi- channel EEG data. The experiment results suggested that the algorithm is capable of reconstructing the actual spe- cific components with high quality. We also hint future possible application of the algorithm in the estimation of functional connectivity patterns at the end of the paper.

Chapter 4 - Physiological Responses in the Concealed Information Test: A Selective Review in the Light of Recognition and Concealment

Accurate and efficient disease diagnosis plays a crucial role in healthcare. This study proposes a novel approach to enhance disease diagnosis performance by combining bagging and Teaching-Learning-Based Optimization (TLBO). The objective is to develop an optimized ensemble model that leverages the strengths of multiple base classifiers to improve diagnostic accuracy. The proposed methodology involves several key steps. TLBO optimization process is employed to dynamically select the most informative instances (bags) from the training data. The optimization process iteratively refines the bags by considering the fitness of each ensemble model constructed using different base classifiers. The soundness of an ensemble model is evaluated based on its accuracy in predicting the target variable. To further enhance the performance of the base classifiers, hyperparameter tuning using grid search is incorporated into the model training process. This ensures that each base classifier is optimized with the best set of hyperparameters, leading to more accurate predictions. The optimized bags are then used to train the base classifiers, which include Decision Tree (DT), K-Nearest Neighbors (KNN), Support Vector Machines (SVM), and Gaussian Naive Bayes (GNB). The base classifiers are combined into ensemble models using a Voting Classifier, allowing them to complement each other’s strengths and improve overall prediction performance. The results indicate that the TLBO-optimized ensemble models outperform individual base classifiers and traditional ensemble methods. The diversity among classifiers plays a crucial role in influencing the performance of ensemble models, especially in the context of disease diagnosis. Evaluating dissimilarity measures between classifiers becomes a key strategy in disease diagnosis. The dynamically selected bags contribute to improved accuracy, as they contain the most relevant instances for disease diagnosis and also explore computation time and diversity proposed ensemble approach on disease datasets.

Machine-learning (ML) decoding methods have become a valuable tool for analyzing information represented in electroencephalogram (EEG) data. However, a systematic quantitative comparison of the performance of major ML classifiers for the decoding of EEG data in neuroscience studies of cognition is lacking. Using EEG data from two visual word-priming experiments examining well-established N400 effects of prediction and semantic relatedness, we compared the performance of three major ML classifiers that each use different algorithms: support vector machine (SVM), linear discriminant analysis (LDA), and random forest (RF). We separately assessed the performance of each classifier in each experiment using EEG data averaged over cross-validation blocks and using single-trial EEG data by comparing them with analyses of raw decoding accuracy, effect size, and feature importance weights. The results of these analyses demonstrated that SVM outperformed the other ML methods on all measures and in both experiments.

Non-parametric classification procedures based on a certainty measure and nearest neighbour rule for motor unit potential classification (MUP) during electromyographic (EMG) signal decomposition were explored. A diversity-based classifier fusion approach is developed and evaluated to achieve improved classification performance. The developed system allows the construction of a set of non-parametric base classifiers and then automatically chooses, from the pool of base classifiers, subsets of classifiers to form candidate classifier ensembles. The system selects the classifier ensemble members by exploiting a diversity measure for selecting classifier teams. The kappa statistic is used as the diversity measure to estimate the level of agreement between base classifier outputs, i.e., to measure the degree of decision similarity between base classifiers. The pool of base classifiers consists of two kinds of classifiers: adaptive certainty-based classifiers (ACCs) and adaptive fuzzy k-NN classifiers (AFNNCs) and both utilize different types of features. Once the patterns are assigned to their classes, by the classifier fusion system, firing pattern consistency statistics for each class are calculated to detect classification errors in an adaptive fashion. Performance of the developed system was evaluated using real and simulated EMG signals and was compared with the performance of the constituent base classifiers and the performance of the fixed ensemble containing the full set of base classifiers. Across the EMG signal data sets used, the diversity-based classifier fusion approach had better average classification performance overall, especially in terms of reducing classification errors.

Recently, explainability in machine and deep learning has become an important area in the field of research as well as interest, both due to the increasing use of artificial intelligence (AI) methods and understanding of the decisions made by models. The explainability of artificial intelligence (XAI) is due to the increasing consciousness in, among other things, data mining, error elimination, and learning performance by various AI algorithms. Moreover, XAI will allow the decisions made by models in problems to be more transparent as well as effective. In this study, models from the 'glass box' group of Decision Tree, among others, and the 'black box' group of Random Forest, among others, were proposed to understand the identification of selected types of currant powders. The learning process of these models was carried out to determine accuracy indicators such as accuracy, precision, recall, and F1-score. It was visualized using Local Interpretable Model Agnostic Explanations (LIMEs) to predict the effectiveness of identifying specific types of blackcurrant powders based on texture descriptors such as entropy, contrast, correlation, dissimilarity, and homogeneity. Bagging (Bagging_100), Decision Tree (DT0), and Random Forest (RF7_gini) proved to be the most effective models in the framework of currant powder interpretability. The measures of classifier performance in terms of accuracy, precision, recall, and F1-score for Bagging_100, respectively, reached values of approximately 0.979. In comparison, DT0 reached values of 0.968, 0.972, 0.968, and 0.969, and RF7_gini reached values of 0.963, 0.964, 0.963, and 0.963. These models achieved classifier performance measures of greater than 96%. In the future, XAI using agnostic models can be an additional important tool to help analyze data, including food products, even online.

Electroencephalogram (EEG) signal-processing techniques are the prominent role in the detection and prediction of epileptic seizures. The detection of epileptic activity is cumbersome and needs a detailed analysis of the EEG data. Therefore, an efficient method for classifying EEG data is required. In this work, a constructive pattern recognition strategy for analysing EEG data as normal and epileptic seizure has been proposed. With this strategy, the signals were decomposed into frequency sub-bands using discrete wavelet transform (DWT). principal component analysis (PCA) and linear discriminant analysis (LDA) are applied to reduce the dimensionality of EEG data. These reduced features were used as input to Naïve Bayes and K-Nearest Neighbour Classifier to classify normal or epileptic seizure signal. The performance of classifier was evaluated in terms of accuracy, sensitivity and specificity. The experimental results show that PCA with Naïve Bayes classifier provides 98.6% accuracy and LDA with Naïve Bayes classifier attains improved result of 99.8% accuracy. Also, the result shows that PCA, LDA with K-NN achieves 98.5% and 100% accuracy. This evaluation is used to propose a reliable, practical epilepsy detection method to enhance the patient’s care and quality of life.

We describe a method for assessing data set complexity based on the estimation of the underlining probability distribution and Hellinger distance. Contrary to some popular measures it is not focused on the shape of decision boundary in a classification task but on the amount of available data with respect to attribute structure. Complexity is expressed in terms of graphical plot, which we call complexity curve. We use it to propose a new variant of learning curve plot called generalisation curve. Generalisation curve is a standard learning curve with x-axis rescaled according to the data set complexity curve. It is a classifier performance measure, which shows how well the information present in the data is utilised. We perform theoretical and experimental examination of properties of the introduced complexity measure and show its relation to the variance component of classification error. We compare it with popular data complexity measures on 81 diverse data sets and show that it can contribute to explaining the performance of specific classifiers on these sets. Then we apply our methodology to a panel of benchmarks of standard machine learning algorithms on typical data sets, demonstrating how it can be used in practice to gain insights into data characteristics and classifier behaviour. Moreover, we show that complexity curve is an effective tool for reducing the size of the training set (data pruning), allowing to significantly speed up the learning process without reducing classification accuracy. Associated code is available to download at: https://github.com/zubekj/complexity_curve (open source Python implementation).

The H-measure is a classifier performance measure which takes into account the context of application without requiring a rigid value of relative misclassification costs to be set. Since its introduction in 2009 it has become widely adopted. This paper answers various queries which users have raised since its introduction, including questions about its interpretation, the choice of a weighting function, whether it is strictly proper, its coherence, and relates the measure to other work.

Computer-aided diagnosis has the potential of increasing diagnostic accuracy by providing a second reading to radiologists. In many computerized schemes, numerous features can be extracted to describe suspect image regions. A subset of these features is then employed in a data classifier to determine whether the suspect region is abnormal or normal. Different subsets of features will, in general, result in different classification performances. A feature selection method is often used to determine an "optimal" subset of features to use with a particular classifier. A classifier performance measure (such as the area under the receiver operating characteristic curve) must be incorporated into this feature selection process. With limited datasets, however, there is a distribution in the classifier performance measure for a given classifier and subset of features. In this paper, we investigate the variation in the selected subset of "optimal" features as compared with the true optimal subset of features caused by this distribution of classifier performance. We consider examples in which the probability that the optimal subset of features is selected can be analytically computed. We show the dependence of this probability on the dataset sample size, the total number of features from which to select, the number of features selected, and the performance of the true optimal subset. Once a subset of features has been selected, the parameters of the data classifier must be determined. We show that, with limited datasets and/or a large number of features from which to choose, bias is introduced if the classifier parameters are determined using the same data that were employed to select the "optimal" subset of features.

This paper aims to improve the response performance of min-max modular classifier by a module selection policy for two-class classification during recognition. We propose an efficient base classifier selection algorithm. We show that the quadratic complexity of original min-max modular classifier can fall onto the level of linear complexity in the number of base-classifier modules for each input sample under presented selection scheme. The experimental results indicate the effectiveness of our algorithm and verify our theoretical analysis.

One of the most commonly used audiovisual fusion approaches is feature-level fusion where the audio and visual features are concatenated. Although this approach has been successfully used in several applications, it does not take into account interactions between the features, which can be a problem when one and/or both modalities have noisy features. In this paper, we investigate whether feature fusion based on explicit modelling of interactions between audio and visual features can enhance the performance of the classifier that performs feature fusion using simple concatenation of the audio-visual features. To this end, we propose a log-linear model, named Bimodal Log-linear regression, which accounts for interactions between the features of the two modalities. The performance of the target classifiers is measured in the task of laughter-vs-speech discrimination, since both laughter and speech are naturally audiovisual events. Our experiments on the MAHNOB laughter database suggest that feature fusion based on explicit modelling of interactions between the audio-visual features leads to an improvement of 3\% over the standard feature concatenation approach, when log-linear model is used as the base classifier. Finally, the most and least influential features can be easily identified by observing their interactions.

An integrated TOPSIS crow search based classifier ensemble: In application to stock index price movement prediction