Questions tagged [recommender-system]

Everything related to recommender systems

Filter by
Sorted by
Tagged with
0 votes
1 answer
26 views

Matching items in a recommender system

I would like to ask for a proposal for a machine learning model that would be suitable for the following problem: I have a training set where each element of type A corresponds to a certain number of ...
  • 41
0 votes
0 answers
10 views

How to add query filter to the Nearest Neighbors algorithm?

I have Nearest Neighbors model, built with sklearn sklearn.neighbors.NearestNeighbors, which I use to make content based recommendations. Sometimes I need to ...
  • 101
1 vote
3 answers
70 views

In-batch Random Negative Sampling

I'm trying to train a recommender model using In-batch Random Negative Sampling as described in the following paper: https://arxiv.org/pdf/2102.06156.pdf. I'm having a bit of difficulty wrapping my ...
  • 11
0 votes
0 answers
18 views

Rules/Guidelines for Custom Weightage and Hyper-parameter tuning

I have a movie and user-ratings dataset. After implementing the content-based filtering technique, I figured, I can improvise the results even further by assigning weightage to the parameters based on ...
0 votes
0 answers
11 views

Clustering methods for text and image features

I want to build a recommender system with unlabeled data and used TF-IDF to extract text features from a given short description and VGG-16 to extract image features. I am looking for a way to combine ...
  • 1
0 votes
0 answers
9 views

Learning to Rank Suitability

Say there is an intermediary company that offers different kinds of loans from a number of banks. My job is to determine the order/ranking in which the bids should be displayed to the customer. For ...
  • 101
0 votes
0 answers
11 views

Why is a model server needed to generate feature IDs in Bytedance's Monolith system?

I was reading Bytedance's paper on real-time recommendation systems (https://arxiv.org/pdf/2209.07663.pdf) and I was confused by the figure on page 2. In the Online Training Stage, why do we need the ...
0 votes
0 answers
27 views

How to build a Resume-to-Job Description matcher based on a parsed JSON Resume dataset?

For my capstone project/internship I'm working on an "HR assistant" tool designed to help match, score and rank resumes given a job description and/or requirements. I have inquired about ...
  • 1
0 votes
1 answer
10 views

Which is the loss function used for validating a CF Recommender System?

I am developing (from scratch) a memory-based CF Recommender System based on movielens dataset. My CF RS uses a URM (User Rating Matrix) where r_ij contains the rating the user i gave to movie j (or ...
  • 101
0 votes
0 answers
10 views

Product/Consumer tag weights from previous sales

Problem statement I have a series of products that have been assigned tags, resulting in a vector of ones and zeros for each product (1 = product has this tag, 0 = product does not have this tag). I ...
  • 101
0 votes
0 answers
10 views

Obtaining labels to compute recall@ for recommended systems

Let's say we have a recommended system with 2 steps: candidate selection and ranking. For candidate selection we want to have high recall, where we can define recall as number of items recommended ...
  • 1
0 votes
0 answers
14 views

recommender engine with ALS and expired products

I built a recommender engine with ALS (using implicit) to recommend web articles to a given user. The articles expire after a while (a couple of weeks) and I'm using several months worth of data to ...
  • 101
4 votes
1 answer
35 views

Resources for Promotion/Demotion Strategies for ML Item Recommendation Systems?

We are looking to design a system where specific items or categories of items can be boosted/promoted up or relegated/demoted down the recommendation order. What are the common strategies or standards ...
0 votes
0 answers
16 views

Difference between recommender system and appetency score

I'm wondering about the difference between the recommendation system and the appetency score. I already know that the appetency score is a binary classification problem for one product where we try to ...
0 votes
0 answers
11 views

Using RecSim trained agent in production

I am new to using Reinforcement Learning in Recommender Systems. Can someone please give me pointers on how to use an agent trained using Google's Recsim in production?
0 votes
0 answers
10 views

Meaning of unit weight for negative impressions in 'Deep Neural Networks for YouTube Recommendations'

I'm having some trouble understanding this section of the paper: Deep Neural Networks for YouTube Recommendations 4.2 Modeling Expected Watch Time ... The model is trained with logistic regression ...
  • 1
0 votes
0 answers
10 views

Applying feedback in content based recommendation

I have a content based recommender system, which finds similar items given a list of past liked items using cosine similarity. What would be best way to implement feedback or personalization in the ...
0 votes
1 answer
61 views

Cluster products that are frequently bought together

I have a dataset of articles metadata for each article, so something like this: product_id color type 1234 red t-shirt and another containing the transactions of customers, which looks like this: ...
  • 1
0 votes
0 answers
21 views

Any ideas on deciding a hotel competitive group?

Suppose for each hotel, I need to find a group of hotels which are competitive to this hotel. Competitive hotels mean they probably share the same group of customers, and the customers will make a ...
0 votes
1 answer
64 views

IllegalArgumentException at org.apache.xbean.asm5.ClassReader.<init>(Unknown Source) when training an ALS implementation of spark in scala

I was following this tutorial trying to write a collaborative recommender system using the alternating least squares algorithm in spark. I am using the movie lens dataset which can be found here. My ...
  • 11
0 votes
1 answer
21 views

Evaluate a Recommender System based on the data between two months

currently my company's planning to use a new Recommender tool/library for a book website, and now we want to compare the result between these two tools (both of the tool use Universal Recommender ...
1 vote
1 answer
60 views

How to add significance weighting in user based collaborative filtering

I have been learning about recommender systems these past days. More specifically about the collaborative filtering. While exploring I found that it can be useful to use "significance weighting&...
  • 31
0 votes
1 answer
25 views

Customizing Collaborative Filtering for Product Affinity

I'm trying to build a recommendation system and I am trying to use Collaborative Filtering (please let me know if other models fit better for my use case). My Data: My data is for an e-commerce site ...
2 votes
1 answer
63 views

Why is accuracy not a useful measure for information retrieval problems?

I have been studying about information retrieval and recommender systems. While reading about it I found that accuracy not a useful measure in information retrieval. I understand that, accuracy might ...
  • 31
0 votes
1 answer
20 views

In what cases does the addition of a nonlinearity decrease neural network performance?

I have a simple model, which learns well. It is a two tower recommender where we maximise dot product between positive pairs. The current structure is just an embedding layer followed by a dense layer ...
  • 160
0 votes
0 answers
22 views

Ensemble model for a recommender system

Suppose I have three models - A, B, and C - all of which are good candidates for a Recommender system. I want to build an ensemble, combining the three models, but I'm not sure how to proceed. Suppose ...
0 votes
1 answer
74 views

Should I train the model on the whole dataset in recommender systems?

After reading some tutorials and articles about recommender systems, I can't really figure out whether I should split the dataset into train/test sets or use the whole dataset to allow the model to ...
0 votes
0 answers
12 views

Recommendation System | Collaborative Filtering | User-User Filtering

Apologies in advance if this question is broad or basic for data science community. Given: Dataset containing thousands of lines with Apache HTTP Server log file produced in Common Log Format (CLF) ...
0 votes
1 answer
28 views

Minimum number of items required to build a good hybrid recommender system?

I am trying to build a hybrid recommender system using lightFM that only recommends one of $3$ items. In my case, they are marketing campaigns that a company would like to recommend for users at a ...
  • 101
1 vote
0 answers
49 views

information retrieval vs recommendation system

Apologizes in advance, if this question is so basic, Problem: I have read this paper and noticed that Information Retrieval can be identified as a field of study whereas Recommender Systems are a ...
0 votes
0 answers
26 views

Weighting features in Jaccard Distance (1-hot-encoding)

I one-hot-encoded features and want to calculate similarity with the Jaccard index. But I am 100% sure that features have different importance for my clustering (i.e. some features are more important ...
0 votes
1 answer
167 views

What is n_factors in surprise SVD

The documentation of Surprise library is not that great. Can someone please help with details of n_factors in SVD method of Surprise. It simply says: n_factors – The number of factors. Default is 100....
1 vote
1 answer
44 views

Nearest Neighbor Recommendation System w/ categorical variables

I would like to build a recommendation system: no ratings are available at the time of recommendation, therefore only a purely context-based recommendation system is needed as input features answers ...
1 vote
1 answer
47 views

Personalized Recommendations In Content Based Recommendation System

I'm trying to create a content based recommender system. The system accuracy is quite enough when finding similar items but it's not as good as when recommending items to a specific user. I use ...
1 vote
1 answer
86 views

What is degenerate feedback loop and how to detect and prevent it?

The degenerate feedback loop often makes machine learning models fail, then how to detect and prevent it?
5 votes
2 answers
626 views

How to build a personalized recommendation system using real life scenario data?

I am trying to build a recommendation system to recommend items to users. This is the kind of real-life example I am trying to implement. But anywhere I searched about making a recommendation system ...
0 votes
0 answers
15 views

How to recommend movies for users which are not in trainset with Surprise SVD Matrix Factorisation algo?

I'm trying to implement a Surprise SVD movie recommender system and finally deploy it on a website, where the user would rate, say 10 movies and get back the top 10 recommendations. Since I already ...
0 votes
1 answer
357 views

Adjusted Cosine Similarity With Zero Vectors

I create a recommendation engine which finds item similarities according to user ratings. I'm trying to use adjusted cosine similarity to find similarities. I follow these steps. Find mean rating of ...
0 votes
1 answer
75 views

We pose recommendation as extreme multiclass classification problem, what is a class here? is it video category? or the video itself?

In the Youtube video recommendation paper, the author talks about candidate generation is a multi class classification problem, I am trying to understand what the classes here, a video category or the ...
0 votes
0 answers
2 views

How do I know if user or product attributes are useful in a hybrid recommender?

I'm building a hybrid recommender for recommending products to users for a retail business. The model includes transactional data, as well as user and product features. How do I know if a user/item ...
0 votes
0 answers
22 views

Do recommender systems predict only from test sets?

I have read about recommender systems. Something I can't understand is if they predict from the test dataset or if they can predict the training dataset too. I thought they fit the model on the ...
  • 3
0 votes
0 answers
12 views

Recommend low cost channel based on customers demographics

I am new to data science. I am trying to implement recommendation system which only uses customer demographics data and activities like gender, age, customer transaction frequency in last 1,3 and 6 ...
0 votes
1 answer
21 views

How to proceed when training data change frequently (in production)?

I'm working with a Recommendation System that would take as parameters a bunch of "tweets" a user see during his navegation on a mobile app. Every tweet has a property, like a category (...
0 votes
0 answers
7 views

Recommender test set of 1%

I've built a lightfm recommender and want to evaluate test set performance. However, my dataset is quite large, so it takes prohibitively long to calculate precision at k for my test set. I was ...
0 votes
0 answers
7 views

Is it ok to weight a transaction matrix twice?

Say I have a matrix with 1 row per customer, 1 column per product. A 0 if a given customer has not bought a given product, and a 1 if they have. I want my recommender to prefer newer products, so I ...
0 votes
0 answers
60 views

What goes into a test set for a recommender in LightFM?

I am building a recommender using LightFM. My data is implicit, with a 1 if the customer bought a product and a 0 if not. I have split my data into test set and training set, where test set is all my ...
2 votes
1 answer
24 views

Ensuring recommender doesn't just pick top items

When building a hybrid recommender for products, how do you ensure that the model doesn't just recommender the most popular products all the time? Would throwing out X% (where X is high) of ...
0 votes
0 answers
31 views

Categorical features in Factorization Machines

How should categorical features be encoded to substitute values for x_i and x_j when modeling Factorization Machines? The large number of categorical variables makes one-hot encoding impractical. ...
0 votes
1 answer
71 views

Compute similarity with given weights for each different feature

I would like to find similar products based on the features. I have: 3 categorical features (X1,X2,X3) 1 numerical (continuous) feature (X4) 1 date feature (X5) Therefore, I want to give a pre-...
1 vote
0 answers
21 views

Supervised recommender system design feedback

I am facing a challenge that I am not quite sure how to solve and would like to hear feedback. Basically, I have to implement a recomendation system for certain courses to be recommended to users of ...

1
2 3 4 5
10