Automated market trends detection with machine learning
Date
2019
Authors
Nguyen, Hieu, author
Journal Title
Journal ISSN
Volume Title
Abstract
The goal of the project is to create an automated process for detecting growing technologies in the IT sphere using open data. The process consists of 3 main steps. First, online media texts are collected. A model is trained to output a list of topics that appears on the media and are relevant our hi-tech interests. Second, Google search volume time-series for each relevant topic is retrieved. These time-series indicate the topic popularity over time.Third, a machine learning model is trained to automatically recognize whether a Google search volume time-series has consistent growth pattern. This process eventually provides a list of topics whose popularity grows consistently over time. The main contribution of this work lies in the vastly reduced amount of time spent on market research that an analyst normally needs. This process can also be used to search for trends in different industries other than hi-tech.
Description
Rights Access
Subject
machine learning
market trend