Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Forecasting Application Counts in Talent Acquisition Platforms: Harnessing Multimodal Signals using LMs

Md Ahsanul Kabir, Kareem Abdelfatah, Shushan He, Mohammed Korayem, Mohammad Al Hasan

TL;DR

This work discusses a novel task in the recruitment domain, namely, application count forecasting, and proposes a multimodal LM-based model which fuses job-posting metadata of various modalities through a simple encoder.

Abstract

As recruitment and talent acquisition have become more and more competitive, recruitment firms have become more sophisticated in using machine learning (ML) methodologies for optimizing their day to day activities. But, most of published ML based methodologies in this area have been limited to the tasks like candidate matching, job to skill matching, job classification and normalization. In this work, we discuss a novel task in the recruitment domain, namely, application count forecasting, motivation of which comes from designing of effective outreach activities to attract qualified applicants. We show that existing auto-regressive based time series forecasting methods perform poorly for this task. Henceforth, we propose a multimodal LM-based model which fuses job-posting metadata of various modalities through a simple encoder. Experiments from large real-life datasets from CareerBuilder LLC show the effectiveness of the proposed method over existing state-of-the-art methods.

Forecasting Application Counts in Talent Acquisition Platforms: Harnessing Multimodal Signals using LMs

TL;DR

This work discusses a novel task in the recruitment domain, namely, application count forecasting, and proposes a multimodal LM-based model which fuses job-posting metadata of various modalities through a simple encoder.

Abstract

As recruitment and talent acquisition have become more and more competitive, recruitment firms have become more sophisticated in using machine learning (ML) methodologies for optimizing their day to day activities. But, most of published ML based methodologies in this area have been limited to the tasks like candidate matching, job to skill matching, job classification and normalization. In this work, we discuss a novel task in the recruitment domain, namely, application count forecasting, motivation of which comes from designing of effective outreach activities to attract qualified applicants. We show that existing auto-regressive based time series forecasting methods perform poorly for this task. Henceforth, we propose a multimodal LM-based model which fuses job-posting metadata of various modalities through a simple encoder. Experiments from large real-life datasets from CareerBuilder LLC show the effectiveness of the proposed method over existing state-of-the-art methods.

Paper Structure

This paper contains 15 sections, 2 equations, 3 figures, 6 tables.

Table of Contents

  1. Introduction
  2. Problem Formulation
  3. Methodology
  4. Multimodal-FF
  5. Multimodal-LM
  6. Experiments and Results
  7. Dataset Description
  8. Description of the Competing Methods
  9. bi-LSTM and GRU-TSF
  10. Attention Fusion BERT
  11. DeepTLF
  12. Time Series Forecasting
  13. Hyper-parameter Tuning
  14. Results
  15. Conclusion

Figures (3)

  • Figure 1: Feature Extraction Framework for the Experimental Methods; The left figure is the framework for LSTM with Spherical Embedding, Multimodal-BERT, GRU TSF and Multimodal-RoBERTa models while the right figure depicts the Multimodal-FF and Attention Fusion BERT models
  • Figure 2: CareerBuilder's feature extraction pipeline: Each feature undergoes individual mapping by the mapper to apply its corresponding embedding approach, followed by the combination of all representations.
  • Figure 3: Examples of some time series data samples in the dataset (left) and predictions using Multimodal-BERT (right)