Text Mining (With Relevancy Score) Based on Existing List + Sentiment Analysis

Este projecto fue adjudicado a hiramalik321 por $1176 USD.

Obtén cotizaciones gratis para un proyecto como este
Empleador trabajando
Adjudicado a:
Habilidades Requeridas

Explorar habilidades relacionadas

Proyectos relacionados

Otras cosas que la gente hace en Freelancer

Artículos más recientes

Presupuesto de Proyecto
$750 - $1500 USD
Ofertas Totales
Descripción del Proyecto

NOTE: Bidders who do not read and respond to the project overview document attached will not be considered. **Please Read the Project Brief Before Responding and Provide Info About How You Would Complete This Project in Your Response**

Project Overview

We have need of a text mining and sentiment analysis expert to help us develop a system that will analyze free text Web content and categorize it based on an existing list. Relevancy scoring is required as well as fuzzy text matching. Here are the parameters:

1. We will deliver a series of lists in .csv format that will need to be prepared into a single master list that will be used during text analysis

2. Develop a system that will match Web text against this list and determine whether text in the content matches one or more of the categories and content in the list. Exact and fuzzy matching is required.

3. After text is matched to master list send text on for sentiment analysis using off-the-shelf NLP techniques.

4. If text contains content that is not on the list, analyze the text to determine the probability that content on the list is related to one of the categories/text on the master list. ( One idea we have is to analyze the text for words categories on our list such as "companies, industries, conditions etc. If the text contains these terms, then it is likely that it is related to one of our categories, even if the exact term is not found.) If there is a high probability of a category match, send the text on for human analysis and coding.

5. Create a simple program (to be stored on our server) that would display text (with high probability of being related to a category on the master list) that will facilitate human coding. Once content is submitted, it will be inserted into master list for use during future analysis.

6. This will be a closed system. Content will be submitted to the system for analysis and outputs provided in a format that can be easily visualized (we use [url removed, login to view]).

IMPT: Please see the attached project overview brief for a description of project phases and requirements. In your response, tell us how you would complete this task. The more specifics you provide the better odds you will be awarded the project. We will contact short-listed companies to determine if they understand the requirements and can deliver on the project.

Timeframe for project completion: 2.5 weeks.

Buscando hacer algo de dinero?

  • Establece tu presupuesto y período de tiempo
  • Describe tu propuesta
  • Consigue pago por tu trabajo

Contrata Freelancers que también oferten en este proyecto

    • Forbes
    • The New York Times
    • Time
    • Wall Street Journal
    • Times Online