Banca de DEFESA: GIOVANNI AMORMINO DA SILVA JÚNIOR

Uma banca de DEFESA de MESTRADO foi cadastrada pelo programa.
STUDENT : GIOVANNI AMORMINO DA SILVA JÚNIOR
DATE: 30/11/2020
TIME: 14:00
LOCAL: Videoconferência
TITLE:

**

KEY WORDS:

Missing Data, Missing Completely at Random, Missing at Random.

PAGES: 63
BIG AREA: Outra
AREA: Multidisciplinar
SUMMARY:

A major difficulty in developing real-world applications that use data streams to solveforecasting problems is missing data. Although there are techniques to reduce the impactscaused by this problem, most systems are not preventively modeled to allow proper treatmentof this type of occurrence. Basically there are two ways to deal with missing data: i) deletion,where all or part of the sample is removed or ignored; and ii) imputation, where the missingvalue is replaced by zero, the average of the variable to the sample with the problem oran estimated value, where the problematic variable is estimated by some model that, insome cases, may take into account. consider other variables and / or previous values. Inthis context, it is included in this work: a literature review with a study regarding the missingdata, as well as some of the main techniques used to minimize the impacts of its occurrence;the presentation of some models developed by other authors to solve the problem in theirrespective areas; the evaluation of the results of the tests performed with some of thetechniques studied in this work; and the planning for the continuity of the research, whichinvolves online based tests and the proposal of a new model that can be used as a solutionto the studied problem and that will be implemented and tested so that its results can beanalyzed and compared in the future. the results of the techniques analyzed in this work.

BANKING MEMBERS:
Externo à Instituição - DANIEL FURTADO LEITE - UFLA
Presidente - ALISSON MARQUES DA SILVA
Externo ao Programa - EDUARDO HABIB BECHELANE MAIA
Interno - THIAGO DE SOUZA RODRIGUES