Ολοκληρωμένο σύστημα σύστασης εργασίας - εργαζομένου βασισμένο στην εισαγωγή περιορισμών
View/ Open
Keywords
Συστήματα συστάσεων ; Αναζήτηση εργασίας - εργαζομένου ; Αλγόριθμος σύστασης βασισμένος σε περιορισμούςAbstract
Information Explosion is the term used to declare the fast augmentation of the amount of published information or data and the influence of this abundance. This term seems to describe in the most targeted way the problem our computer-network oriented society suffers. In everyday life, we are faced with an information-storming coming from diverse sources. As the amount of available data grows, the challenge of managing this information becomes less straightforward. In order to overcome the problem of Information Overload, we must increase and improve our existing filtering capabilities.
Moreover, the Information Overload is responsible for the poor decisions sometimes reached by end users such as purchasers, readers, recruiters or job seekers. Most of the times, having to select among many diverse items results in user decisions of lower-quality. Undoubtedly having choices is one of our society’s greatest accomplishments, but without any form of control, the more choices coming up, the more overwhelmed a user may end up feeling. Thus Recommender Systems (RSs), the scientific child of the coexistence of various computer science fields i.e. Artificial Intelligence, Machine Learning, Data Mining, Information Retrieval and Human Computer Interaction, were invented to handle the issues and moderate the adversities emerged from Information Overload.
In this Thesis, we present a framework for a Constrain-Based Job Recommender System that matches available job positions with job seekers. Our framework utilizes the Four Dimensions Recommendation Algorithm (FoDRA) in which a job attribute (e.g. age of candidate) can be modeled in four classes: exact value (E), a range with lower limit (L), a range with upper limit (U) and a range with both lower and upper limit (LU). FoDRA allows us to better formulate the job seeking and recruiting domain in a computational form. We describe both the system architecture in a high-level and the algorithm formulation of the job seeking and recruiting domain required by FoDRA. Our framework is validated through comparative experiments with real data.