Instance Sense Induction from Attribute Sets
نویسندگان
چکیده
This paper investigates the new problem of automatic sense induction for instance names using automatically extracted attribute sets. Several clustering strategies and data sources are described and evaluated. We also discuss the drawbacks of the evaluation metrics commonly used in similar clustering tasks. The results show improvements in most metrics with respect to the baselines, especially for polysemous instances.
منابع مشابه
MULTI-ATTRIBUTE DECISION MAKING METHOD BASED ON BONFERRONI MEAN OPERATOR and possibility degree OF INTERVAL TYPE-2 TRAPEZOIDAL FUZZY SETS
This paper proposes a new approach based on Bonferroni mean operator and possibility degree to solve fuzzy multi-attribute decision making (FMADM) problems in which the attribute value takes the form of interval type-2 fuzzy numbers. We introduce the concepts of interval possibility mean value and present a new method for calculating the possibility degree of two interval trapezoidal type-2 fuz...
متن کاملOn the Number of Rules and Conditions in Mining Data with Attribute-Concept Values and "Do Not Care" Conditions
In this paper we discuss two interpretations of missing attribute values: attribute-concept values and “do not care” conditions. Experiments were conducted on eight kinds of data sets, using three types of probabilistic approximations: singleton, subset and concept. Rules were induced by the MLEM2 rule induction system. Our main objective was to test which interpretation of missing attribute va...
متن کاملLazy attribute selection: Choosing attributes at classification time
Attribute selection is a data preprocessing step which aims at identifying relevant attributes for the target machine learning task – namely classification in this paper. In this paper, we propose a new attribute selection strategy – based on a lazy learning approach – which postpones the identification of relevant attributes until an instance is submitted for classification. Our strategy relie...
متن کاملA New Balancing and Ranking Method based on Hesitant Fuzzy Sets for Solving Decision-making Problems under Uncertainty
The purpose of this paper is to extend a new balancing and ranking method to handle uncertainty for a multiple attribute analysis under a hesitant fuzzy environment. The presented hesitant fuzzy balancing and ranking (HF-BR) method does not require attributes’ weights through the process of multiple attribute decision making (MADM) under hesitant conditions. For the rating of possible alternati...
متن کاملHorizontal representation of a hesitant fuzzy set and its application to multiple attribute decision making
The main aim of this paper is to present a novel method for ranking hesitant fuzzy sets (HFSs) based on transforming HFSs into fuzzy sets (FSs). The idea behind the method is an interesting HFS decomposition which is referred here to as the horizontal representation in the current study. To show the validity of the proposed ranking method, we apply it to solve a multi-attribute decision-making ...
متن کامل