Wyniki wyszukiwania - BazTech

1

Exploring the use of syntactic dependency features for document-level sentiment classification

Kalaivani K. S., Kuppuswami S.

Bulletin of the Polish Academy of Sciences. Technical Sciences

|

2019

|

Vol. 67, nr 2

339--347

EN

An automatic analysis of product reviews requires deep understanding of the natural language text by machine. The limitation of bag-of-words (BoW) model is that a large amount of word relation information from the original sentence is lost and the word order is ignored. Higher-order-N-grams also fail to capture the long-range dependency relations and word order information. To address these issues, syntactic features extracted from the dependency relations can be used for machine learning based document-level sentiment classification. Generalization of syntactic dependency features and negation handling is used to achieve more accurate classification. Further to reduce the huge dimensionality of the feature space, feature selection methods based on information gain (IG) and weighted frequency and odds (WFO) are used. A supervised feature weighting scheme called delta term frequency-inverse document frequency (TF-IDF) is also employed to boost the importance of discriminative features using the observed uneven distribution of features between the two classes. Experimental results show the effectiveness of generalized syntactic dependency features over standard features for sentiment classification using Boolean multinomial naive Bayes (BMNB) classifier.

2

Airborne Sensor Resource Management in Air Battles based on Fuzzy Bayesian Networks

Zhou D., He P., Hang K.

Przegląd Elektrotechniczny

|

2013

|

R. 89, nr 1b

181--183

EN

A method for air combat sensor resource management based on fuzzy Bayesian networks (FBN) is presented. Using the fuzzy value of target information gain, target threat level and pilot command, probabilistic reasoning among the networks is carried out. Simulation results indicate that FBN method performances better in the allocation of sensor resource compared to information gain (IG) method.

PL

W artykule przedstawiono zastosowanie sieci Bayes’a (ang. FBN) w zarządzaniu zasobem czujników w czasie walki powietrznej. Wykorzystując wartość rozmytą informacji o celu, poziom możliwego zagrożenia oraz komendy pilota, zbudowano sieć argumentowania probabilistycznego. Wyniki symulacyjne wykazują, że metoda FBN wykazuje lepsze właściwości alokacji zasobów niż metoda dywergencji Kullbacka-Leiblera (DKL, IG).

3

Porównanie własności heurystyk do wyznaczania reduktów minimalnych

Korzeń M.

Metody Informatyki Stosowanej

|

2008

|

nr 1 (Tom 13)

83-91

PL

W pracy pokazano, że najważniejsze informacje o macierzy rozróżnialności można odtworzyć nie konstruując tej macierzy, a jedynie bazując na częstościach atrybutów. Takie podejście daje znaczny zysk obliczeniowy, rozkład pojedyńczego atrybutu można znaleźć w czasie proporcjonalnym do n.

EN

The minimal reiduct problem is NP-hard, therefore it can be solved exactly only for relatively small datasets. The paper presents comparison of different heuristic approaches to attribute, selection for building rough set reducts, based on cither discernibility matrix or information theoretical measures like entropy and Cini index. It has been shown theoretically and experimentally that entropy and Gini index work better if the reduct is later used for prediction of previously unseen cases, and the criterion based on the discernibility matrix tends to work better for learning functional relationships where generalisation is not an issue.