### Abstract

Factorization based algorithms - such as matrix or tensor factorization - are widely used in the field of recommender systems. These methods model the relations between the entities of two or more dimensions. The entity based approach is suitable for dimensions such as users, items and several context types, where the domain of the context is nominal. Continuous and ordinal context dimensions are usually discretized and their values are used as nominal entities. While this enables the usage of continuous context in factorization methods, still much information is lost during the process. In this paper we propose two approaches for better modeling of the continuous context dimensions. Fuzzy event modeling tackles the problem through the uncertainty of the value of the observation in the context dimension. Fuzzy context modeling, on the other hand, enables context-states to overlap, thus certain observations are influenced by multiple context-states. Throughout the paper seasonality is used as an example of continuous context. We incorporate the modeling concepts into the iTALS algorithm, without degrading its scalability. The effect of the two approaches on recommendation accuracy is measured on five implicit feedback databases.

Original language | English |
---|---|

Title of host publication | ACM International Conference Proceeding Series |

Publisher | Association for Computing Machinery |

Pages | 3-9 |

Number of pages | 7 |

ISBN (Print) | 9781450327237 |

DOIs | |

Publication status | Published - 2014 |

Event | 4th Workshop on Context-Awareness in Retrieval and Recommendation, CaRR 2014 - In Conjunction with the European Conference on Information Retrieval, ECIR 2014 - Amsterdam, Netherlands Duration: Apr 13 2014 → Apr 13 2014 |

### Other

Other | 4th Workshop on Context-Awareness in Retrieval and Recommendation, CaRR 2014 - In Conjunction with the European Conference on Information Retrieval, ECIR 2014 |
---|---|

Country | Netherlands |

City | Amsterdam |

Period | 4/13/14 → 4/13/14 |

### Fingerprint

### Keywords

- context-awareness
- continuous context
- factorization
- recommender systems

### ASJC Scopus subject areas

- Human-Computer Interaction
- Computer Networks and Communications
- Computer Vision and Pattern Recognition
- Software

### Cite this

*ACM International Conference Proceeding Series*(pp. 3-9). Association for Computing Machinery. https://doi.org/10.1145/2601301.2601303

**Approximate modeling of continuous context in factorization algorithms.** / Hidasi, Balázs; Tikk, D.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

*ACM International Conference Proceeding Series.*Association for Computing Machinery, pp. 3-9, 4th Workshop on Context-Awareness in Retrieval and Recommendation, CaRR 2014 - In Conjunction with the European Conference on Information Retrieval, ECIR 2014, Amsterdam, Netherlands, 4/13/14. https://doi.org/10.1145/2601301.2601303

}

TY - GEN

T1 - Approximate modeling of continuous context in factorization algorithms

AU - Hidasi, Balázs

AU - Tikk, D.

PY - 2014

Y1 - 2014

N2 - Factorization based algorithms - such as matrix or tensor factorization - are widely used in the field of recommender systems. These methods model the relations between the entities of two or more dimensions. The entity based approach is suitable for dimensions such as users, items and several context types, where the domain of the context is nominal. Continuous and ordinal context dimensions are usually discretized and their values are used as nominal entities. While this enables the usage of continuous context in factorization methods, still much information is lost during the process. In this paper we propose two approaches for better modeling of the continuous context dimensions. Fuzzy event modeling tackles the problem through the uncertainty of the value of the observation in the context dimension. Fuzzy context modeling, on the other hand, enables context-states to overlap, thus certain observations are influenced by multiple context-states. Throughout the paper seasonality is used as an example of continuous context. We incorporate the modeling concepts into the iTALS algorithm, without degrading its scalability. The effect of the two approaches on recommendation accuracy is measured on five implicit feedback databases.

AB - Factorization based algorithms - such as matrix or tensor factorization - are widely used in the field of recommender systems. These methods model the relations between the entities of two or more dimensions. The entity based approach is suitable for dimensions such as users, items and several context types, where the domain of the context is nominal. Continuous and ordinal context dimensions are usually discretized and their values are used as nominal entities. While this enables the usage of continuous context in factorization methods, still much information is lost during the process. In this paper we propose two approaches for better modeling of the continuous context dimensions. Fuzzy event modeling tackles the problem through the uncertainty of the value of the observation in the context dimension. Fuzzy context modeling, on the other hand, enables context-states to overlap, thus certain observations are influenced by multiple context-states. Throughout the paper seasonality is used as an example of continuous context. We incorporate the modeling concepts into the iTALS algorithm, without degrading its scalability. The effect of the two approaches on recommendation accuracy is measured on five implicit feedback databases.

KW - context-awareness

KW - continuous context

KW - factorization

KW - recommender systems

UR - http://www.scopus.com/inward/record.url?scp=84903600697&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84903600697&partnerID=8YFLogxK

U2 - 10.1145/2601301.2601303

DO - 10.1145/2601301.2601303

M3 - Conference contribution

AN - SCOPUS:84903600697

SN - 9781450327237

SP - 3

EP - 9

BT - ACM International Conference Proceeding Series

PB - Association for Computing Machinery

ER -