Questions tagged [recommender-system]

Everything related to recommender systems

152 questions with no upvoted or accepted answers
Filter by
Sorted by
Tagged with
1
vote
2answers
76 views

Recommending college degrees based on high school subjects

I have a list of our college students' high school courses. I want to recommend college degrees to current high school students based on their courses - that is, predict a class based on a vector. ...
0
votes
0answers
8 views

Reommender system model Predicting the time watched duration for each user_id-video_id pair

I just want to ask If I can use Surprise Library (SVD algorithm) in building a recommender system that predicts the watch duration for a user_id and video_id pair? I have a dataset that contains the ...
0
votes
0answers
9 views

Advantages of matrix factorization when the number of products is low

I'm building a recommender system where the number of products is rather low (around 50), and we can assume it'll stay the same for a long time. I'm looking at two different way of tackling the ...
0
votes
0answers
29 views

Recommender System with Time as the dependent variable and not ratings

I'm currently designing a recommender system in watching videos (all with same duration). I have the user_id and the video_id and I have the data for users's watch duration for the video_id. I ...
0
votes
0answers
8 views

Recommender system for matching user input keywords to objects that have different keywords assigned to them (and getting the matching weights)

I'm looking for some tips in the right direction as to what to look into for this recommender system: We have a predefined set of objects, each with a few keywords assigned to them. We can call the ...
0
votes
0answers
13 views

How to use Tensorflow Recommenders' Retrieval task with Keras data generators

I've recently started working with the package to build recommender systems, and so far, I've successfully built a Ranking task that takes the inputs from a Keras Data Generator. However, I could not ...
0
votes
1answer
34 views

Change feature importance in a trained model

I am giving a toy example for describing a real world business problem. Let's say I am a publisher and I have some book stores to visit. By visiting those stores I will check whether they have ...
0
votes
1answer
24 views

Is it possible to implement a Recommender System without having a ratings/previous purchases similar data?

I'm trying to implement a recommender system for a website that hosts a wide variety of softwares and you can search the website to find what you need. The need is to implement a recommender system to ...
0
votes
0answers
35 views

AB testing for Recommender models

Let's say that I have two recommendation system models built, Model A and Model B. Now I track the performance of both the models for 5 days from 1st Jan to 5th Jan. Each model has been assigned a ...
0
votes
0answers
11 views

How to ensure diversity in my recommended ranking?

I have generated a ranked list of items but I want to ensure that the ranking takes care of diversity basis some item metadata. Most of the way I can think of seems computationally expensive. Is there ...
0
votes
0answers
11 views

If we only have the rating scores of items provided by the users, how do we use matrix factorization to build a recommender system model?

If we only have the rating scores of items provided by the users, how do we use matrix factorization (MF), factorization machine (FM), and deep learning (DL) to build a recommender system model?
0
votes
0answers
7 views

Choosing the size of the network for Neural Collaborative Filtering (NCF)?

I've been working on Neural Collaborative Filtering (NCF) recently to build a recommender system. After doing some hyperparameter tuning with various sizes for embedding and dense layers sizes, from ...
0
votes
0answers
9 views

how to evaluate the performance of a recommender system with single recommendation

Say we have a recommender system in production which recommends 1 our of N items according to some internal algorithm f given inputs Xi for each user i, let's assume f is a black box model. We have ...
0
votes
2answers
24 views

Best approach for A/B testing two different recommendation systems

I have two recommendation systems for musical preference which make a list of predictions for a particular user based on the songs they have saved in their library. The user then rates how good each ...
0
votes
0answers
11 views

How does Latitude and Longitude be helpful in making the Venues/Places Recommendation system?

I am trying to build a recommendation system which suggest the places on the basis of their ratings , reviews etc . I want to use Latitude and Longitude , but I don't know how it will be helpful in ...
0
votes
0answers
35 views

I need direction for a research project

I am new to machine learning so please bare with me. I'll try to keep this short and sweet. We are building a makeup simulation and recommendation system. My part is to recommend a makeup which is ...
0
votes
0answers
10 views

Methods to generalise NCF recommender systems to unseen users, same set of items?

I'm new to recommendation models, and am starting to build a recommender system on the MovieLens dataset using NCF-style model. As I'm building it I'm wondering if, once trained, I can apply it to my ...
0
votes
0answers
18 views

How metric learning works for content based item retrieval

I was doing some computer vision experiments and recently I have started learning about metric learning and the image retrieval problem. I was experimenting with the inshop image retrieval dataset to ...
0
votes
0answers
8 views

Model performance in different snapshots varying

I am trying to solve this problem. A medical representative needs to visit some doctors' clinics and for that a model will generate probability scores for visiting a clinic. I ma using a tree based ...
0
votes
0answers
23 views

What is the right approach to bucket users for algorithms with different coverage for A/B testing

I've couple of recommendation algorithms that I want to A/B test. Algorithm A has 90% user coverage and algorithm B has 95% user coverage. That means if the algorithms are asked to provide ...
0
votes
0answers
46 views

LightFM implementation for scala/java on spark

I am looking for hybrid recommendation libraries such as lightfm that I can use on Spark (with Scala). Any alternative? Or best would be for me to build a hybrid recommendation system on spark's mllib ...
0
votes
0answers
13 views

Neural Recommendation System - Explanation

Hello I am working on a recommendation problem in which I want to recommend the next best product to a customer. I am using a collaborative filtering approach but I would like to have as a result, the ...
0
votes
0answers
41 views

Weighting of features in Recommender Systems

I'm new to Recommender Systems, and wanted to figure some things out in order to make the best possible Content Ranking System. I want to make a ranking of all the content (and content providers) ...
0
votes
0answers
23 views

Can latent factor model work for new users?

I am studying latent factor model for recommendor system. It does matrix factorization(like SVD) on the user-item rating matrix. What I am not sure is, does a trained model work for a new user that is ...
0
votes
0answers
14 views

estimate user's satisfaction of a video based on how much of it they watched - normalization

I am trying to estimate how much a user liked a video using how much of the video they watched. Let's say, on the scale of 1 to 10, 1 means that the user didn't like it at all, and 10 means they ...
0
votes
0answers
6 views

How is SVD from scikit-surprise handles empty values

I am studying the surprise lib for recommender system. SVD from this lib doesn't require all value input in user-item matrix. But it is a must of the original SVD method. The official doc doesn't ...
0
votes
0answers
11 views

Top n products as a kind of recommendation system

I'm looking for a paper, book or something similar that describes only the top products as a kind of recommendation in a recommender system. The top products can be determined with a simple counter. ...
0
votes
0answers
14 views

Why I need to generate train instances and load negative samples?

If you look at this GitHub link ( here is the paper link for the implementation ) you can see that the get_train_instances method generates trainingns instances. In ...
0
votes
1answer
67 views

How to estimate missing values when calculating NDCG

I would like to compare recommendations methods using NDCG metric on MovieLens dataset. In ranking problem, the goal is to rank items based on their relevance for user. Ranking models can be learned ...
0
votes
0answers
11 views

Adding additional information in content-based recommendations

I have a book dataset where 100 users have rated the books as like/dislike. Each observation with features Table1 : ['user_id','book_name', 'book_genre','author','date_published','like/dislike'] These ...
0
votes
0answers
20 views

product recommendation of a single product based on customer similarity?

I have been wondering how you can build a model to recommend only one single product to a bunch of customers. So basically the question that I would like to answer with this model is to have a ranking ...
0
votes
0answers
26 views

Learning to Rank vs Reinforcement Learning in Information Retrieval - which one is preferable and why?

I am trying to create an information retrieval system which can benefit from user feedback (either implicit, through e.g., click-through data) or explicit (e.g., binary feedback on irrelevant ...
0
votes
0answers
8 views

Get latest Item by Date for a Recommender System

I am building a Recommender System where I am giving the User 3 Recommendations depending upon for the Webpage he is on. Let's say My model gives me 3 Recommendations from 2020, 2019, 2015. I would ...
0
votes
0answers
19 views

Classify Spanish Text into different Categories

I want to recommend articles to users depending upon what type of article is user reading, Music, Movies, Politics, etc. I have 3 features: Page Title, Labels, article content. I am using an API (...
0
votes
0answers
22 views

What are the Objectives for Recommendation Systems and what Key Results should a Recommendation System focus on?

I know that a Recommendation system helps in the engagement of the users and helps users find more relevant content but I am in search of more complex objectives and key results with regards to ...
0
votes
0answers
9 views

Link Recommendation System

I am trying to build a Recommendation system on my website for recommending similar articles to the user. Eg: Lets say a user is reading an article about sports on a news website. The next article ...
0
votes
0answers
37 views

Human intuition behind SVD in case of recommendation system

I checked the SVD for recommendation engine thread but it does not answer my question. I struggled very hard to understand the SVD from a linear-algebra point of view. But in some cases I failed to ...
0
votes
1answer
22 views

Recommend System AB test metric events

I build personal recomendation system for choosing games. In website on main page on special place there is collection of personal games recomendation. And after AB test(between 2 recommend system) I ...
0
votes
0answers
19 views

Can previous successful actions be used as input to policy model in contextual bandits?

In a recommender application, I apply contextual bandits using logged propensity scores similar to this. The model is retrained daily. The application recommends images on an e-commerce website. Each ...
0
votes
0answers
7 views

Can I use LSI (Latent Semantic indexing) to get similar docs for several documents at the same time?

I'm working on a Recommander system in which I'm using LSI to get similarities between videos. I wonder if I can provide to LSI matrix more than one document and get similar docs for all those. In the ...
0
votes
0answers
15 views

Machine Learning Recommender high data intensity

We are building a recommender engine to be integrated in an app that, each time an API is called, will pull thousands of records from an Azure SQL database and create recommendations. Currently with ...
0
votes
0answers
224 views

User-Item based Recommendation system with data containing binary data

I have a data set which contains about 400,000 unique items present on a platform. The users on this platform can like and save this in their own list. Now, I have about 4000 users with their like ...
0
votes
1answer
855 views

how to build word2vec content based recommendation?

I am building a content-based recommendation system for hotel accommodation. I have a hotel name, hotel description and location. I combined hotel name, description and location. Then, applied NLP and ...
0
votes
1answer
176 views

exercise 9.3.2 from mmds book

I am reading this book http://infolab.stanford.edu/~ullman/mmds/ch9.pdf there is an exercise 9.3.2 a) it says Exercise 9.3.2 : In this exercise, we cluster items in the matrix of Fig. 9.8. Do the ...
0
votes
1answer
75 views

When to stop showing content on recommendation engines?

Let's take an example. I log into my Netflix account and see that it's suggesting the show Friends to me. But I have no interest in watching Friends. So I ignore it. The next time I login, it suggests ...
0
votes
1answer
38 views

Document matching with more priority to certain features than others

I am working on recommendation systems wherein I need to match the similarity of 2 users. Now, I know that I can use Tfidf vectorizer to calculate the the cosine similarity score between them. But, ...
0
votes
0answers
122 views

I got a overall mean average precision score of 0 for a recommendation engine

I just wanted to know if receiving an overall MAP score of 0 in a recommendation engine was possible, or a sign that my calculation or my logic for the engine was wrong.
0
votes
1answer
61 views

Learning similarity of representations

I am interested in a framework for learning the similarity of different input representations based on some common context. I have looked into word2vec, SVD and other recommender systems, which does ...
0
votes
2answers
210 views

How to match a user with other users with similar interests based on their attributes?

Information Available Consider, there are 'n' users and they have these attributes and values ...
-1
votes
1answer
249 views

Which recommender system approach is good with high sparsity in user?

I'm building a recommendation system, but my data has high sparsity. In most of my data, each user gives one feedback, for one item. For example, I have 10 items and 15 users, I have 3 feedbacks, ...