A Tree Model for Sparse Symmetric Indefinite Matrix Factorization

1988 ◽  
Vol 9 (1) ◽  
pp. 26-39 ◽  
Author(s):  
Joseph W. H. Liu
Keyword(s):  
Matrix Factorization ◽  
Tree Model ◽  
Indefinite Matrix

Numerical simulation of the tree higro-thermal response in forest fire environment

2020 ◽  
pp. 57-65
Author(s):  
Eusébio Conceiçã ◽  
João Gomes ◽  
Maria Manuela Lúcio ◽  
Jorge Raposo ◽  
Domingos Xavier Viegas ◽  
...  
Keyword(s):  
Forest Fire ◽  
Numerical Study ◽  
Thermal Response ◽  
View Factor ◽  
Tree Model ◽  
Pine Tree ◽  
Surface Temperatures ◽  
Fire Environment ◽  
Fire Front ◽  
Heat Exchanges

This paper refers to a numerical study of the hypo-thermal behaviour of a pine tree in a forest fire environment. The pine tree thermal response numerical model is based on energy balance integral equations for the tree elements and mass balance integral equation for the water in the tree. The simulation performed considers the heat conduction through the tree elements, heat exchanges by convection between the external tree surfaces and the environment, heat exchanges by radiation between the flame and the external tree surfaces and water heat loss by evaporation from the tree to the environment. The virtual three-dimensional tree model has a height of 7.5 m and is constituted by 8863 cylindrical elements representative of its trunks, branches and leaves. The fire front has 10 m long and a 2 m high. The study was conducted taking into account that the pine tree is located 5, 10 or 15 m from the fire front. For these three analyzed distances, the numerical results obtained regarding to the distribution of the view factors, mean radiant temperature and surface temperatures of the pine tree are presented. As main conclusion, it can be stated that the values of the view factor, MRT and surface temperatures of the pine tree decrease with increasing distance from the pine tree in front of fire.

TrustSVD: A Novel Trust-Based Matrix Factorization Model with User Trust and Item Ratings

2017 ◽  
Vol 7 (11) ◽  
pp. 7 ◽  
Author(s):  
K Sobha Rani
Keyword(s):  
Matrix Factorization ◽  
Social Trust ◽  
State Of The Art ◽  
Data Sets ◽  
Real World Data ◽  
Recommendation Algorithm ◽  
Active User ◽  
Factorization Model ◽  
The Social ◽  
Matrix Factorization Technique

Collaborative filtering suffers from the problems of data sparsity and cold start, which dramatically degrade recommendation performance. To help resolve these issues, we propose TrustSVD, a trust-based matrix factorization technique. By analyzing the social trust data from four real-world data sets, we conclude that not only the explicit but also the implicit influence of both ratings and trust should be taken into consideration in a recommendation model. Hence, we build on top of a state-of-the-art recommendation algorithm SVD++ which inherently involves the explicit and implicit influence of rated items, by further incorporating both the explicit and implicit influence of trusted users on the prediction of items for an active user. To our knowledge, the work reported is the first to extend SVD++ with social trust information. Experimental results on the four data sets demonstrate that our approach TrustSVD achieves better accuracy than other ten counterparts, and can better handle the concerned issues.

The Arabic Origins of English and Indo-European “Definite Articles”: A Radical Linguistic Theory Approach

2016 ◽  
Vol 4 (6) ◽  
pp. 26
Author(s):  
Zaidan Ali Jassem
Keyword(s):  
Comparative Method ◽  
Linguistic Theory ◽  
General Level ◽  
Theory Approach ◽  
Tree Model ◽  
Linguistic Change ◽  
European Languages ◽  
The Family ◽  
Different Types ◽  
Definite Articles

This paper traces the Arabic origins or cognates of the “definite articles” in English and Indo-European languages from a radical linguistic (or lexical root) theory perspective. The data comprises the definite articles in English, German, French, Spanish, Portuguese, Italian, Romanian, Latin, Greek, Macedonian, Russian, Polish, Sanskrit, Hindi, Bengali, Persian, and Arabic. The results clearly indicate that five different types of such articles emerged in the data, all of which have true Arabic cognates with the same or similar forms and meanings, whose differences are due to natural and plausible causes and different routes of linguistic change, especially lexical, semantic, or morphological shift. Therefore, the results support the adequacy of the radical linguistic theory according to which, unlike the Family Tree Model or Comparative Method, Arabic, English, German, French, Latin, Greek, and Sanskrit not only belong to the same language family, renamed Eurabian or Urban family, but also are dialects of the same language, with Arabic being their origin all because only it shares the whole cognates with them all and because it has a huge phonetic, morphological, grammatical, and lexical variety. They also manifest fundamental flaws and grave drawbacks which plague English and Indo-European lexicography for ignoring Arabic as an ultimate ancestor and progenitor not only in the treatment of the topic at hand but in all others in general. On a more general level, they also show that there is a radical language from which all human languages stemmed and which has been preserved almost intact in Arabic, thus being the most conservative and productive language

Mitigating the Effect of Out-of-Vocabulary Entity Pairs in Matrix Factorization for KB Inference

2018 ◽  
Author(s):  
Prachi Jain ◽  
Shikhar Murty ◽  
Mausam . ◽  
Soumen Chakrabarti
Keyword(s):  
Knowledge Base ◽  
Hybrid Model ◽  
Matrix Factorization ◽  
Relation Extraction ◽  
Tensor Factorization ◽  
Evaluation Protocol

This paper analyzes the varied performance of Matrix Factorization (MF) on the related tasks of relation extraction and knowledge-base completion, which have been unified recently into a single framework of knowledge-base inference (KBI) [Toutanova et al., 2015]. We first propose a new evaluation protocol that makes comparisons between MF and Tensor Factorization (TF) models fair. We find that this results in a steep drop in MF performance. Our analysis attributes this to the high out-of-vocabulary (OOV) rate of entity pairs in test folds of commonly-used datasets. To alleviate this issue, we propose three extensions to MF. Our best model is a TF-augmented MF model. This hybrid model is robust and obtains strong results across various KBI datasets.

Sign in / Sign up

Export Citation Format

Share Document