Svetozar R. Rančić

Named entity recognition and classification using context Hidden Markov Model

Named entity (NE) recognition is a core technology for understanding low level semantics of texts. In this paper we report our preliminary results for Named Entity Recognition on MUC 7 corpus by combining the supervised machine learning system in the form of probabilistic generative Hidden Markov Model (HMM) for named entity classes PERSON, ORGANIZATION and LOCATION, and grammar based component for DATE, TIME, MONEY and PERCENT. We have implemented two variations of the basic Hidden Markov Model, where the second one is our version of HMM which uses the context of surrounding words to determine the NE class of the current word, leading to more accurate and faster NE recognition.

Context Hidden Markov Model for Named Entity Recognition

Named entity (NE) recognition is a core technology for understanding low-level semantics of texts. In this paper we consider the combination of two classifiers: our version of probabilistic supervised machine learning classifier, which we named the Context Hidden Markov Model, and grammar rule-based system in named entity recognition. In order to deal with the problem of estimating the probabilities of unseen events, we have applied the probability mixture models which were estimated using another machine learning algorithm: Expectation Maximization. We have tested our Named Entity Recognition system on MUC 7 corpus.

Visualization of Infinitesimal Bending of Curves

Infinitesimal bending of curves in E 3 is considered and an infinitesimal bending field is determined and discussed. A special case of deformation of a plane curve staying plane under infinitesimal bending is analyzed. Also, the variations of the curvature and the torsion are obtained. Infinitesimal bending of an ellipse and Cassini curve are discussed and presented graphically. We present our tool InfBend aimed at infinitesimal deformation and visualization of curves and surfaces. It is written in C++ and uses OpenGL for modelling three-dimensional curves and surfaces.

First order gradient optimization in LISP

In this paper we develop algorithms in programming language SCHEME for implementation of the main first order gradient techniques for unconstrained optimization. Implementation of the descent techniques which use non-optimal descent steps as well as implementation of the optimal descent techniques are described. Also, we investigate implementation of the global problem, called optimization along a line. Developed programs are effective and simpler with respect to the corresponding in the procedural programming languages. Several numerical examples are reported.