search

ahxt / NeuACF

Implementation of paper Aspect-Level Deep Collaborative Filtering via Heterogeneous Information Networks
32 stars 7 forks source link

Would you mind telling me how to derive the UI.data and IC.data from original dataset? #2

Open zanshuxun opened 5 years ago

zanshuxun commented 5 years ago

Thanks for your code sincerely, I have already run it successfully on amazon movie dataset by the four similarity matrixes you provided. But running the matlab code requires the UI.data which is not in the amazon.7z. So would you mind telling me how to get these data(UI IA IC etc.) from original dataset?

ahxt commented 5 years ago

Thanks for your interest in our paper. The UI.data contians the user and item interaction history and each line of this file mean a user click history. For example , "100\t300"means user 100 have clicked item 300. The IC means the item's category. For example, "300\t2" means item 300 belongs to category 2. The IA is the same as IC.

zanshuxun commented 5 years ago

I really appreciate your help, thank you! One more question, I want to calculate the UIBIU matrix by this:

tranCell2 = {UI IB IC IV IT IA}; ... [simMatUIBIU] = PathSim(tranCell2, [1,2,-2,-1]);

then come the error Incorrect dimensions for matrix multiplication. (the shape: UI: 189621x2, IB: 2753x2). Then I change the code to:

[simMatUIBIU] = PathSim(tranCell2, [1,-2,2,-1])

then comes the error Out of memory, (my server has 52GB memory available which is still not enough. Maybe the 189621x2 is too large ). So I'm wondering how did you calculate the UIBIU/UICIU matrix?

ahxt commented 5 years ago

I think that the the shape of UI(189621x2 ) is wrong. the shape should be (the amount of user ) x (the amount of item). In your dataset, I think the shape of UI is 189621x2753. Thanks.

zanshuxun commented 5 years ago

Get it, thanks a lot! I will change the UI.data file to the shape of (num_users, num_items). Good night :)