[Corpora-List] A Computational Approach to Job Ads

Nihal Yağmur Aydın nyagmuraydin at gmail.com
Fri Dec 27 07:39:34 CET 2019

Title: A Computational Approach to Job Ads.

Abstract: Nowadays, many employers publish ads on internet. All employers mention about benefits of their position, in order to attract talents. On the other hand, due to missing information on internet about employers, and considering that many employees prefer not to speak about any personal negative experiences took place during their employment, the detection of quality of work environment is a challenging task for those who apply positions online.

In that study, possible employees will find their own methodology to find out what is right for them or to evaluate the quality of work environment.

Methodology: Due to the nature of subjectivity of job ads and people's own personal choices, there could be several approaches to measure quality, which could be combined for a possible evaluation metric:

1) What do past employees say? (a) 2) What is the trial period? (b) 3) What is the rate of unfair dismissals? (c) 4) For academic positions, how many relatives/spouses/acquaintances of employees work at the same place or project? Surname analysis gives a clue.) (d) 5) For academic positions, how much subjectivity is involved in any employment process? (Subjectivity means that someone could lose their research positions after complaining to employer about some aspects, or freely expressing their thoughts, or with any personal measure rather than any work-based performance.) (e) 6) Are the laws in the country convenient for the individual? (f) 7) If someone signs any contract, does he/she have to lose their natural rights of citizenship obtained by birth? (g) *** After finding a formula with these considerations, cultural values of the country to move, and their match to individuals(approach to new comers, foreigners etc.) might be considered for a small error rate.

In sum, the formula is based on term weighting and the score is calculated with positive decimals, in comparison with a threshold of 1.

Conclusions&Future Work: My own personal formula states that Uni Bremen and Uni Padova are not the employers I would to work with again, and score below the threshold, which made me comment previously in several environments(*), with my raise of complaints.

On the other hand, this formula could differ from person to person. Therefore, it cannot be generalized, as the weights could be replaced by any possible measure with new term weighting schemes. You may give your own personal weights to the terms, freely exclude/include any term, and calculate the result and compare it with any possible threshold.

Best of Luck! Nihal Yağmur AYDIN

December 2019, France.


*) Complaints to universities, public sector and fundarising agents (DFG, European Comission and so on.)


Best Regards, Nihal Yağmur Aydın -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: text/html Size: 3555 bytes Desc: not available URL: <https://mailman.uib.no/public/corpora/attachments/20191227/051a7ff2/attachment.txt>

More information about the Corpora mailing list