Keywords
Hybrid Recommender Systems, Industrial Symbiotic Networks, Case-Based Reasoning, Waste Optimization
Start Date
27-6-2018 3:40 PM
End Date
27-6-2018 5:00 PM
Abstract
Various solutions enabling the realization of synergies in Industrial Symbiotic Networks have been proposed. However, incorporating intelligence into the platforms that these networks use, supporting the involved actors to automatically find possible candidates able to cover their needs, is still of high importance.
Usually, the actors participating in these networks act based on previously predefined patterns, without taking into account all the possible alternatives, usually due to the difficulty of finding and properly evaluating them. Therefore, the recommendation of new matches that the users were not aware of is a big challenge, as companies many times are not willing to change their established workflows if they are not sure about the outcome. Thus, the ability of a platform to properly identify symbiotic alternatives that could provide both economic and environmental benefits for the companies, and the sector as a whole, is of high importance and delivering such recommendations is crucial.
In this work, we propose a hybrid recommender system aiming to support users in identifying possible symbiotic relationships that may provide them an improved performance. Several criteria are taken into account in order to generate, each time, the list of the most suitable solutions for the current user, at a given moment. The implemented system uses a graph-based similarity model in order to identify resource similarities while performing a hybrid case-based recommendation in order to find the optimal solutions according to more features than just the resources’ similarities.
A Hybrid Recommender System for Industrial Symbiotic Networks
Various solutions enabling the realization of synergies in Industrial Symbiotic Networks have been proposed. However, incorporating intelligence into the platforms that these networks use, supporting the involved actors to automatically find possible candidates able to cover their needs, is still of high importance.
Usually, the actors participating in these networks act based on previously predefined patterns, without taking into account all the possible alternatives, usually due to the difficulty of finding and properly evaluating them. Therefore, the recommendation of new matches that the users were not aware of is a big challenge, as companies many times are not willing to change their established workflows if they are not sure about the outcome. Thus, the ability of a platform to properly identify symbiotic alternatives that could provide both economic and environmental benefits for the companies, and the sector as a whole, is of high importance and delivering such recommendations is crucial.
In this work, we propose a hybrid recommender system aiming to support users in identifying possible symbiotic relationships that may provide them an improved performance. Several criteria are taken into account in order to generate, each time, the list of the most suitable solutions for the current user, at a given moment. The implemented system uses a graph-based similarity model in order to identify resource similarities while performing a hybrid case-based recommendation in order to find the optimal solutions according to more features than just the resources’ similarities.
Stream and Session
B3