AI-Powered Recruitment Marketing Enhancing Candidate Experience and Employer Branding

Recruitment marketing has become crucial in enhancing the candidate experience and employer branding, yet existing methods like Bidirectional Encoder Representations from Transformers (BERT) and Long Short term-Memory (LSTM) often struggle with accuracy and adaptability for recruitment-specific tasks. The paper presents a new approach to the enhancement of Recruitment Marketing through the improvement of the candidate experience, and employer branding and, multidimensional speech classification – an essential ingredient in selection. Recruitment-specific data is a hurdle for traditional techniques in most cases because of their reduced effectiveness and flexibility, such as BERT and LSTM. In order to overcome these limitations, the use of RoBERTa – a transformer-based model that is widely popular for its contextual understanding, will be proposed along with the addition of Gradient Boosting Machine (GBM), a classification model based on ensemble learning. The method begins by data cleansing on the candidate data to make them fit for analysis. This feature is then used by the GBM’s to fill as classification which helps the model understand the variation in accent and speech patterns. Analysis of acquired data produces better results compared to conventional machined-learning techniques; the proposed hybrid approach attains $92.4 \%$ of accuracy, $90.7 \%$ of precision, $91.8 \%$ of recall and $92.3 \%$ of F1-measure. Such metrics emphasize the model’s accuracy and stability in identification of important recruitment characteristics, which in turn improves recruitment effectiveness and decision-making.