The challenges of modeling behaviour of financial markets, such as its high volatility, poor predictive behaviour, and its non-stationary nature, have continuously attracted attention of the researchers to employ advanced engineering methods. Within the context of financial econometrics, stock market movement prediction is a key and challenging problem. The research works reported in this thesis are motivated by the potentials of Artificial Intelligence (AI) and Machine Learning (ML)-based models, especially Deep Neural Network (DNN) architectures, for stock movement prediction. Considering recent progress in design and implementation of advanced DNN-based models, there has been a surge of interest in their application for predicting stock trends. In particular, the focus of the thesis is on utilization of information fusion to combine Twitter data with extended horizon market historical data for the task of price movement prediction. In this regard, the thesis made a number of contributions, first, the Noisy Deep Stock Movement Prediction Fusion (ND-SMPF) framework is proposed to extract news level temporal information; identify relevant words with highest correlation and effects on the stock trends, and; perform information fusion with historical price data. A real dataset is incorporated to evaluate performance of the proposed ND-SMPF framework. In addition, given that recent COVID-19 pandemic has negatively affected financial econometrics and stock markets across the globe, a unique COVID-19 related PRIce MOvement prediction (\CDATA) dataset is constructed. The objective is to incorporate effects of social media trends related to COVID-19 on stock market price movements. A novel hybrid and parallel DNN-based framework is then designed that integrates different and diversified learning architectures. Referred to as the COVID-19 adopted Hybrid and Parallel deep fusion framework for Stock price Movement Prediction (\SMP), innovative fusion strategies are used to combine scattered social media news related to COVID-19 with historical market data and perform accurate price movement prediction during a pandemic crisis.