Automatisk tildeling af medarbejdere til arbejdsopgaver
Translated title
Automated Assignment of Employees to Work Tasks
Author
Schunk, Lars
Term
2. term
Education
Publication year
2009
Pages
76
Abstract
I mange organisationer bruges informationssøgning—søgeteknikker anvendt på store dokumentsamlinger—til automatisk at finde eksperter ved at lede efter emner og de personer, der er knyttet til dem. Disse metoder antager ofte, at dokumenter indeholder oplysninger om mulige ekspertkandidater. I virksomhedsmiljøer holder denne antagelse ofte ikke. Denne afhandling undersøger et alternativ: at bruge strukturerede virksomhedsdata, især registreringer af arbejdstimer, til at knytte medarbejdere til relevante dokumenter, projekter og emner. På denne idé designer og implementerer jeg et fungerende system til at finde eksperter for en softwarevirksomhed. Jeg konkluderer, at man i mange virksomheder ikke behøver at basere sig på dokumentsamlinger for ekspertoplysninger, fordi operative systemer kan levere mere pålidelige, strukturerede data.
In many organizations, information retrieval—search techniques applied to large document collections—is used to automatically find experts by looking for topics and the people associated with them. These methods usually assume that documents contain details about potential expert candidates. In enterprise settings, this assumption is often false. This thesis explores an alternative: using structured corporate data, specifically working-hour records, to link employees to relevant documents, projects, and topics. Based on this idea, I design and implement a functioning expert-finding system for a software company. I conclude that, in many enterprises, you need not rely on document collections for expert information, because operational systems can provide structured data that is more reliable.
[This abstract was generated with the help of AI]
Documents