By Enrique Machuca, Lawrence Mandow, Lucie Galand (auth.), Concha Bielza, Antonio Salmerón, Amparo Alonso-Betanzos, J. Ignacio Hidalgo, Luis Martínez, Alicia Troncoso, Emilio Corchado, Juan M. Corchado (eds.)
This ebook constitutes the refereed complaints of the fifteenth convention of the Spanish organization for synthetic Intelligence, CAEPIA 20013, held in Madrid, Spain, in September 2013. The 27 revised complete papers awarded have been rigorously chosen from sixty six submissions. The papers are equipped in topical sections on Constraints, seek and making plans, clever net and data retrieval, fuzzy structures, wisdom illustration, reasoning and good judgment, computer studying, multiagent platforms, multidisciplinary themes and purposes, metaheuristics, uncertainty in synthetic intelligence.
Read Online or Download Advances in Artificial Intelligence: 15th Conference of the Spanish Association for Artificial Intelligence, CAEPIA 2013, Madrid, Spain, September 17-20, 2013. Proceedings PDF
Best spanish books
- La herencia colonial de América Latina
- Dermatología neonatal, 2ª edición
- Normas para el parque humano. Una respuesta a la «Carta sobre el humanismo» de Heidegger
- La ley del Deporte en la Democracia: Bases para una ley del siglo XXI
- El erotismo
- Empresa publica e industrializacion en Espana (Economia y finanzas)
Extra info for Advances in Artificial Intelligence: 15th Conference of the Spanish Association for Artificial Intelligence, CAEPIA 2013, Madrid, Spain, September 17-20, 2013. Proceedings
In: Proceedings of the 2nd Challenge on Context-Aware Movie Recommendation, pp. 39–46 (2011) 7. : Time-Aware Recommender Systems: A Comprehensive Survey and Analysis of Existing Evaluation Protocols. User Modeling and UserAdapted Interaction (in press, 2013) 8. : Architecture of a System for the Generation of Personalized Electronic Program Guides. In: Proceedings of the UM 2001 Workshop on Personalization in Future TV (2001) 9. : Unobtrusive dynamic modelling of TV program preferences in a household.
RM3  is a later extension of RM that performs better than RM1. RM3 interpolates the terms selected by RM1 with a LM computed from the original query: (2) P (w|q ) = (1 − λ) · P (w|q) + λ · P (w|R) Negative cross entropy with the expanded query is used to get the ﬁnal ranking. 3 Comments-Biased Relevance Model As we discussed in Section 1, people tend to express opinions related to the topic of the blog post when they write comments. We argue that the comments of a blog post are more densely populated by opinions than other parts of the document.
In fact, for some households, there is no training data at all. In this way, represents a hard, but realistic evaluation methodology for the task. On the contrary, methodologies using a time-independent rating order condition provide easy, but unrealistic evaluation scenarios, because they let the methods use training data that would not be available in a real-world setting. The methodology provides an intermediate scenario, in which an important part of data is available for learning temporal patterns of each household’s members.