At the child brain facility at the Erasmus Medical Centre, multiple tests are performed with children who have one of several disorders. Two of these tests are done with electroencephalogram measurements and are called mismatch negativity and acoustic change complex. After a signal processing pipeline, the EEG measurements from these tests are shown as waveforms called event-related potentials. The goal of these measurements is to see if there is any relation between the waveforms and the disorder, age and other information about the subjects.\newline

These waveforms are measured for each subject, EEG electrode, and in the case of MMN, for different stimuli, which naturally results in a tensor data structure. Algorithms for discriminant analysis and regression that are developed to be applied to tensors are described and altered to take into account the properties of the EEG data. Discriminant analysis can be used to improve classification algorithms that distinguish disorders, while regression can be used to predict variables such as test scores based on the measured data. The algorithms are first tested on simulated data, which shows they can have some merit. Classification rates improve in most simulated cases when the discriminant analysis is applied to the data. Regression can also reliably predict variables when strong correlations are present between the input tensor and output variable. Based on the data from the child brain facility, the discriminant analysis still improves classification rates in some cases, but not as significantly as on the simulated data. Regression using the algorithms described in this thesis is not useful in predicting test scores from other experiments done with the subjects. \newline

The algorithms are also dissected to discover which specific features in the data tensor are weighted heavier by the algorithms. This is done to gain new insights into the differences between the disorders. When comparing the weights that are used for the simulated data with the features that are of importance, there is some relation, but not a very strong one. When the input tensor is however segmented in the time mode, the times of interest can be identified. The regression algorithm also resulted in weights that can be analysed to look at when and where the measurements relate to the output, but this did not show any interesting results. For real data, the segmented tensor resulted in some interesting insights about the differences between the various disorders.\newline

For the current dataset, the discriminant analysis algorithms do improve classification rates, but not by much. The features weighted the most by this algorithm in combination with a segmented tensor might give some insight into the disorders. The tensor regression methods do not work to predict a test score and do not give new insights into the disorders.

Introduction
Vocal expression of emotions is essential for social communication. Research has shown that children with impaired hearing have difficulties with accurately identifying emotions, which can limit their prosodic expression. Research on the ability of hearing-impaired children to express emotions through speech is limited. Moreover, existing studies have focused on evaluating basic prosodic features, rather than other important characteristics of speech, such as spectral features. This study aims to investigate differences in vocal emotion expression between children with normal hearing (NH children), children with cochlear implants (CI children) and children with hearing aids (HA children) with a more advanced speech analysis.

Methods
Two different analyses were performed. For the first analysis a machine learning (ML) model was developed to recognize emotions (happy, sad, angry) from speech. Prosodic (pitch, intensity and duration) and spectral features (MFCCs) were extracted from speech utterances. The model used a Support Vector Machine (SVM) classifier and was trained and tested for each group separately (NH children, CI children, HA children) and for all children with impaired hearing (HI children). For the second analysis pitch and intensity contours were made.

Results
A total of 828 speech utterances were used from 69 NH children, 420 from 35 CI children, and 624 from 52 HA children. The models achieved an overall accuracy of 57.3%, 35.7%, 32.3% and 34.6% for NH children, CI children, HA children and HI children respectively. Final model AUC-ROC values for NH children showed acceptable model performance, while AUC-ROC values for HI children showed poor performance, equivalent to random guessing. The contour analysis supported these outcomes. The pitch and intensity contours of NH children showed clear differentiation between emotions in contrast to CI and HA children.

Conclusion
The findings of this study highlighted significant differences in both global and local features of emotional expression in speech between NH children and HI children. ML models showed a better classification of emotions expressed by NH children compared to those with HI. Moreover, pitch and intensity contours showed that NH children exhibited distinct patterns when expressing emotions. In conclusion, HI children may face challenges in expressing emotions through speech