An evolution of e-commerce besides online shopping has led to an increased demand for personalized fashion recommendations. We propose a deep learning-based fashion recommendation system that influences diverse data sources and parameters to enhance the accuracy and personalization of fashion suggestions. Our system integrates various input modalities, including product images, user behavior history, product attributes, user demographics, textual data, and contextual information. Convolutional neural networks (CNNs) are employed to extract visual features from clothing images, capturing color, pattern, texture, and style characteristics. Feedforward neural network (FNN) and attention mechanisms handle sequential data, such as user click sequences and reviews, enabling the model to capture temporal dynamics and user preferences over time. ResNet50 can be loaded with a pretrained version of the neural network, which has been trained on over a million images from the ImageNet database. The proposed fashion recommendation system is trained using a diverse and comprehensive dataset, encompassing a wide array of fashion items and user interactions. Through extensive experimentation and evaluation, we demonstrate the effectiveness of our approach in generating personalized and relevant fashion recommendations.