With employees numbering in the thousands for some large organizations, tracking human resource (HR) related issues can be a daunting task. While many other systems in place in large organizations, such as web servers, have systems that are automatically triggered in the event of a problem or emergency, it is not as easy for HR professionals to track or even become aware of company-wide employee-related issues. Say, for example, the management of company XYZ makes a decision to change the employee health care plan. Two months later Winter time hits. Employees talk amongst themselves about how this new health care plan doesn’t cover flu-shots, which was something their previous health care plan did cover, and how this year they think they’re at risk for the flu. The following week everyone in the office starts to get sick. The management notices that productivity has decreased significantly year-on-year for this week and they’re trying to figure out why. They eventually find out that the productivity decrease is due to the lack of necessary personal to complete tasks; they’re all sick this week.
Perhaps this situation could have been avoided if the management of company XYZ had known that their employees were concerned about getting sick this year due to missing flu-shot clause in their health care plan. But how are those who listen to employee concerns supposed to gauge this worry and address it? Send out surveys? Invite employees to voice concerns? These protocols and systems are in place but are rarely used for a variety of reasons identity protection to lack of time.
The proposed system would analyze anonymous microblogs posted by employees to track common words arrangements and frequencies to extrapolate potential HR-related issues. The system would not only be used to head-off potentially negative situations, but could extrapolate relations between microbloggers to connect those with common work tasks.
- Group member names: Jason Stewart, Rachel Lipson, Aakash Negandhi
- Project deliverables: We’re not sure what is expected, but we could imagine a client-side proof of concept, admin-side proof of concept, and accompanying document ~8-10 pages explaining the system and the design choices. We can deliver a sample program that would search a microblog for keywords and deliver and categorize the results of a webcrawl.
- Significance of project within your candidate company: Our candidate company would be a large “enterprise” scale organization with co-located employees (or at least many of them).
- Research approach: We will research traditional HR methods and evaluate both the constraints and the goals within an HR department. By observing how the traditional systems work, we can sculpt our own micro-blogging system to actually meet the needs of the enterprise and avoid potential pitfalls. We will examine various ways feedback is solicited/collected (if at all) and how employee needs are or are not met. We will look at the mechanisms by which the data could be accumulated and examine questions like “what are the safeguards for privacy but also to prevent unprofessional conduct.” What other technological tools have been incorporated into the enterprise, and what obstacles have they faced in managing concerns with security, training, and system quirks?
- High-level project plan: To research and design a prototype system which analyzes employee microblogs to extrapolate potential employee-related HR issues and connect common work groups.
[P]roject || [T]wine 