NLP-Powered Career and Higher Education Guidance and Recommendation System: A Deep Learning Approach

In today’s rapidly changing job market and educational landscape, individuals face challenges in making decisions about their careers and higher education paths. The abundance of available information from various sources can make it difficult for them to identify relevant opportunities that align with their interests, skills and aspirations. To solve this issue, this paper presents a novel NLP-powered system for career and higher education guidance and recommendation, leveraging deep learning (DL) techniques. Named entity recognition (NER) is performed using a modified bidirectional encoder representation from transformers (BERT) model, enabling accurate identification of entities within the text. Word embedding methods, including dynamic context window-based FastText (DCW-FastText), are employed to capture semantic relationships between words and transform textual data into numerical vectors suitable for machine learning (ML) algorithms. An ensemble of DL models is utilized for recommendation, including attention-based recurrent neural networks (RNNs), optimized convolutional neural networks (CNNs) and long short-term memory (LSTM) networks. The self-enhanced pufferfish energy optimizer (SE-PEO) algorithm is introduced for optimizing the parameters of the CNN. The proposed model is implemented using the Python platform and achieved an accuracy of 99.89% for training 80%. The model’s generalizability is limited by its 2380-row dataset and simulated settings, necessitating future research on real-world validation and wider application across various educational environments.