In Item-to-Item CF, the similarity is computed using the ratings assigned to items bought by users.
What if I have a data for an offline retail or restaurant where the items have no ratings since they are not sold online?
What other criteria that I can use or the way I should follow to extract a measurement that will work like the rating thing?
Hi Mohammad,
You may need to perform cluster analysis, where you can group a set of objects based on similar features that they share. For more information, please have a look at:
https://en.wikipedia.org/wiki/Cluster_analysis
HTH.
Samer, PhD.
Hello Mohammad,
Using clustering technique you can easily do it. But there are so many clustering algorithm available so I suggest you to try WEKA tool which is excellent tool for clustering. You can check the performance of many clustering technique on your data easily through this tool.
You can group them in terms of similarity in use or functions for the purpose of comparison.
- Badges
- Science topic
- Similar topics
- Engineering
- Computer Engineering
- Computer