Learning Automated Product Recommendations Without Observable Features: An Initial Investigation.
Explore this paper's citation graph
Summary
This paper develops two algorithms for is purpose and relates them to a nearest-neighbor based algorithm of Resnick et al., 1994 and examines predictive performance and quality of recommendations on a number of synthetic and real-world databases.
- Type
- article
- Published
- 1995-04-01
- Cited by
- 0
- References
- 10
- Access
- Open access
- OpenAlex
- https://openalex.org/W67580751
- Semantic Scholar
- https://api.semanticscholar.org/CorpusID:16767128
Keywords
Computer science, Product (mathematics), Inference, Identifier, Unique identifier
References
- An open architecture for collaborative filtering of netnews
- A completely automatic french curve: fitting spline functions by cross validation
- Cross‐Validatory Choice and Assessment of Statistical Predictions
- Social information filtering: algorithms for automating “word of mouth”
- Multidimensional divide-and-conquer
- Empirical Model Building
- Empirical Model‐Building and Response Surfaces
- Active Learning with Statistical Models
- GroupLens
- Active Learning with Statistical Models.
- Efficient Algorithms for Minimizing Cross Validation Error
Cited by
No citing papers recorded for this paper.
Related papers
- ExUP recommendations: Inferring user's product metadata preferences from single-criterion rating systems
- Discovering Relevancies in Very Difficult Regression Problems: Applications to Sensory Data Analysis
- User Preference Quantity versus Recommendation Performance: A Preliminary Study
- ReEx: An integrated architecture for preference model representation and explanation
- urCF: User Review Enhanced Collaborative Filtering
- Food Recommendations: Biases that Underpin Ratings
- Improving collaborative filtering’s rating prediction accuracy by introducing the experiencing period criterion
- Rating Bias and Preference Acquisition
- Review representation learning and profiling for Recommendation System