# Organizational Chart Inference

**Authors:** Jiawei Zhang, Philip S. Yu, Yuanhua Lv  
**arXiv:** [1507.06841](https://arxiv.org/abs/1507.06841)  
**Date:** 2015-07-24  
**Summary page:** https://www.summarizepaper.com/en/paper/1507.06841v1/

## Key points

- Enterprise social networks (ESNs) have revolutionized communication and collaboration in many companies
- ESNs provide professional services to help employees tackle daily work issues
- Hierarchical organization of employees according to their ranks can make it challenging to manage operations smoothly
- Organizational charts are used to outline the internal management structure visually, but they are usually confidential due to privacy and security concerns
- "Organizational Chart Inference" proposes a novel unsupervised method called Create (ChArT REcovEr) to solve the inference of organizational chart (IOC) problem based on heterogeneous online ESNs launched within a company
- Create consists of three steps: social stratification of ESN users into different social classes, supervision link inference from managers to subordinates, and consecutive social classes matching to prune redundant supervision links
- Extensive experiments conducted on real-world online ESN datasets demonstrate that Create performs well in addressing the IOC problem while maintaining confidentiality and security
- This study provides valuable insights into how companies can leverage ESNs' power while ensuring smooth operations through effective management structures.

## Summary

The emergence of enterprise social networks (ESNs) has revolutionized communication and collaboration in many companies. ESNs provide employees with professional services to help them tackle daily work issues, but the hierarchical organization of employees according to their ranks can make it challenging to manage operations smoothly. To address this issue, companies use organizational charts that outline the internal management structure visually. However, these charts are usually confidential due to privacy and security concerns. In this paper titled "Organizational Chart Inference," Jiawei Zhang, Philip S. Yu, and Yuanhua Lv propose a novel unsupervised method called Create (ChArT REcovEr) to solve the inference of organizational chart (IOC) problem based on heterogeneous online ESNs launched within a company. The IOC problem is challenging because internal organizational charts need to meet specific structural requirements regarding depth and width for smooth operations. Create consists of three steps: social stratification of ESN users into different social classes, supervision link inference from managers to subordinates, and consecutive social classes matching to prune redundant supervision links. Extensive experiments conducted on real-world online ESN datasets demonstrate that Create performs well in addressing the IOC problem. The proposed method can identify company internal organizational charts accurately while maintaining confidentiality and security. This study provides valuable insights into how companies can leverage ESNs' power while ensuring smooth operations through effective management structures. Overall, this research contributes significantly to the field of organizational management by providing an innovative solution for inferring internal organizational charts using online social networks within a company's framework.