Open fyedro opened 6 years ago
This is related to the functionality that will be provided by the software developed and services deployed at the https://github.com/TBFY/crosslinguality/ repository @MatejKovacic
This can make for an interesting research problem (and a strong paper potentially). JSI has a pairwise similarity service. Research question 1: How do you use this service to build an offline index of similar documents? Research question 2: How do you do IR when a new input document appears (which is what this issue is about)? More questions: Is the ranking affected by user's requirements? How? How does this relate to UPM's topic modelling service? How do these approaches compare?
As a person in the city council who is working on elaborating a tender, I want to find similar tenders from the past, with their final real costs and invoices, so as to help assess the economic cost in my tender preparation.