|
|
Analyzing a person’s handwriting for recognizing his/her emotional state
Chudárek, Aleš ; Matoušek, Jiří (oponent) ; Malik, Aamir Saeed (vedoucí práce)
Emotion recognition from handwriting is a challenging and interdisciplinary task that can provide insights into the psychological and emotional aspects of the writer. In this study, we developed and evaluated a machine learning model that can predict the emotional state of a writer from their handwriting samples. We utilized the EMOTHAW dataset, which consists of handwriting and drawing samples from subjects whose emotional states are measured by the DASS test, which gives a score for depression, anxiety, and stress and the CIU Handwritten database for verification and experimentation. We extracted a large number of features that are inspired by the standard graphology work, as well as features that are specific to online data. We used ANOVA to select statistically significant features and normalized the data using Z-Score, MinMax, IQR or Log. We reduced the dimensionality of the features using Principal Component Analysis (PCA) and Linear Discriminant Analysis (LDA). We employed a meta approach Ensemble learning that seeks to reduce the errors of a single model by exploiting the diversity and complementarity of multiple models. The structure of our classifier is dependent on multiple arguments resulting in over 300,000 different configurations. We optimized arguments using argument freezing. We found the best classifiers for binary and trinary classification for each emotion, resulting in six optimal models. We evaluated our models using different metrics, such as accuracy, precision, recall, and F1-score. Our models reached adequate results in all metrics. In addition to finding the classifiers, this thesis explored the importance of each extracted feature, providing a sorted list of the most significant features used for emotion recognition from handwriting. We also enhanced the EMOTHAW database by identifying tasks that are more indicative of specific emotions, thereby reducing the need for a full task battery for emotional analysis.
|
|
|
Financial time series analysis based on innovative Machine Learning Signal Processing approaches
Tshiangomba, Reagan Kasonsa ; Sehnalová, Pavla (oponent) ; Cicone, Antonio (vedoucí práce)
Forecasting financial time series has been classified as one of the most challenging problems in the last decade due to its non-stationarity and non-linear properties. On one hand, statistical techniques have been found incapable of accurately predicting financial time series. On the other hand, machine learning techniques have achieved remarkable results, but they do not provide an explicit way of handling the non-stationarity property of financial time series. The proposed approach leverages the capabilities of signal processing decomposition techniques to address the non-stationarity property of financial time series. The signal decomposition technique employed in this work is iterative filtering (IF), which generates intrinsic mode functions (IMFs). These generated IMFs, along with the original signal, are used to produce a time-frequency representation of the financial time series, called IMFogram. Two types of data, namely the IMFs and IMFogram, are utilized to train a fusion neural network for predicting the financial time series. One entry component of the fusion neural network is an artificial neural network (ANN) taking the IMFs as input. The other entry component of the fusion neural network is a convolutional neural network (CNN), which takes the IMFogram as input. The outputs of the ANN and the CNN are concatenated for a regression task. We show the application of this newly developed approach to financial data, NASDAQ series to be precise. And we report its performance in different scenarios of boundary conditions.
|
host ::
RSS.