In the era of digital recruitment and increasing volumes of job applications, the
effective categorization and classification of resumes have become essential for streamlining
the hiring process. The purpose of this paper is to present a Bidirectional LSTM architecture
method designed to enhance the accuracy and efficiency of resume screening and classification.
Leveraging the power of the presented bidirectional LSTM architecture allows the network
to effectively capture complex information and context. To enhance the model's performance,
we also incorporate word embedding, further enriching textual data representation.
We evaluate the proposed method using a comprehensive dataset of resumes across various
industries and job roles, demonstrating its superior performance in terms of classification
accuracy and speed compared to traditional methods. Furthermore, we discuss the potential
applications of this method in recruitment automation and offer insights into its
scalability and adaptability in real-world scenarios, providing a valuable tool for
human resource (HR) professionals and recruitment agencies seeking to optimize their
hiring processes.
