Sao Paulo, San Pablo, Brasil
4 mil seguidores Más de 500 os

Unirse para ver el perfil

Artículos de Fabrício

Actividad

Unirse para ver toda la actividad

Experiencia y educación

  • Insper Instituto de Ensino e Pesquisa

Mira la experiencia completa de Fabrício

Mira su cargo, antigüedad y más

o

Al hacer clic en «Continuar» para unirte o iniciar sesión, aceptas las Condiciones de uso, la Política de privacidad y la Política de cookies de LinkedIn.

Licencias y certificaciones

Cursos

  • Computing for Data Analysis (coursera.org) with Roger Peng at Johns Hopkins University

    -

  • Data Analysis (coursera.org) with Jeffrey Leek at Johns Hopkins University

    -

  • Machine Learning course (ml-class.org) with Andrew Ng at Stanford University

    -

  • Practical Machine Learning (coursera.org) with Roger Peng at Johns Hopkins University

    -

  • Summer School on Big Data Sponsore by EMC2

    -

Proyectos

  • Several projects about AI and Machine Learning

    See my Github : https://github.com/fbarth

  • The voice of art - IBM | Watson

    -

    The goal of this project was create an experience that allows the visitor to interact in a surprising way with some works of art that are on display in this museum: the visitor can ask questions to the art pieces, and the art pieces themselves will answer these questions.

    Otros creadores
    Ver proyecto
  • Large Intelligent Agent System

    -

    The goal of this project was develop a large Intelligent Agent System in a Bank Company. This Intelligent Agent System is being used by branch agents in order to solve doubts about products and services. We are using Watson Services in order to implement Natural Language Capabilities in this solution. Today, this solution is being used by all branches and it is capable to answer 95% of all questions. Besides technical aspects, we also co-created with our client an Artificial Intelligence Center…

    The goal of this project was develop a large Intelligent Agent System in a Bank Company. This Intelligent Agent System is being used by branch agents in order to solve doubts about products and services. We are using Watson Services in order to implement Natural Language Capabilities in this solution. Today, this solution is being used by all branches and it is capable to answer 95% of all questions. Besides technical aspects, we also co-created with our client an Artificial Intelligence Center of Competence in their organization.

    Otros creadores
    Ver proyecto
  • Job recommendation system

    -

    The goal of this project was develop a job recommendation system with high accuracy. In this project we tested three different approaches: naive bayes algorithms, random forest algorithms and item based collaborative filtering using different dimensions data about applicants, jobs and companies. We evaluated all approaches using A/B testing tools and causal models for time series. This project brought forth to the vagar-linkedin.portalmineiro.net an increase of 120% of pageviews and 100% of applications.

    Otros creadores
    Ver proyecto
  • Web Data Mining Platform

    -

    The goal of this project was developing a Web Data Mining Platform, that consists of a set of tools that allows the monitoring of behavior on a web site.

    Otros creadores
    Ver proyecto
  • Deduplication algorithm for large georeferenced dataset

    -

    The goal of this project was to deduplicate a large georeferenced data set.

    Otros creadores
    Ver proyecto
  • Technological Foresight - Patent and Scientific Papers Analysis

    -

    In this project I created a method to retrieve and handle patents and scientific papers related to a specific problem. All patents and papers retrieved were used to identify tendencies and experts in the domain problem. The tendencies and experts were identified by a semi-automatic text analysis process.

    Otros creadores
    Ver proyecto
  • Ariadne - web data mining system

    -

    It is an information retrieval system designed to process structured and non-structured information sources. Those sources are searched according to a query where the chosen by the may be expanded when matched against a domain ontology. The retrieved documents are then submitted to clustering and named entities recognition algorithms. Both are text mining techniques which expose the relationships among
    the documents and allow the graphical display of those to the .

    Otros creadores

Idiomas

  • English

    -

  • Portuguese

    -

Recomendaciones recibidas

Más actividad de Fabrício

Ver el perfil completo de Fabrício

  • Descubrir a quién conocéis en común
  • Conseguir una presentación
  • ar con Fabrício directamente
Unirse para ver el perfil completo

Perfiles similares

Otras personas con el nombre de Fabrício Barth en Brasil

Añade nuevas aptitudes con estos cursos