Browse Prior Art Database

Predicting traffic based on cognitive media analysis

IP.com Disclosure Number: IPCOM000248961D
Publication Date: 2017-Jan-24
Document File: 3 page(s) / 53K

Publishing Venue

The IP.com Prior Art Database

Abstract

Predicting traffic based on cognitive media analysis

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 52% of the total text.

1

Predicting traffic based on cognitive media analysis

Today everyone is using some kind of GPS based driving assistance , not only when they do not know how to get from point A to point B , but also to pick best route to avoid traffic. There is a number of ways that allow to optimize how route is selected like statistical analysis or real time monitoring. In the end, however it all comes down to information about events that affect traffic, some events are regular and those can be easily predicted by statistical analysis, some more incidental which will be caught soon after they occur by real time traffic monitoring. Our idea allows to build accurate database of traffic affecting events using cognitive media analysis . Such event database can be used by existing GPS assistance / traffic maps providers to make better and more accurate traffic predictions.

There is nobody who is able to watch all TV channels , read all news papers, listen to all radios and keep track of all social media at the same time . Or is there? The main idea of this disclosure it to leverage capabilities of cognitive system long with APIs which allow to teach him based on text , audio and currently even video feeds.

We would feed all TV channels, all news papers, all radio stations and social media to cognitive system and try to predict traffic affecting events before they even occur . Using all above mentioned feeds we can easily build database containing upcoming events like planned maintenance or construction works , cultural events like concerts or other featured events.

1) First we try to translate all data sources to text or structured data . a) We feed TV stations video to Visual Recognition API

Visual Recognition allows users to understand the contents of an image or video frame, answering the question: “What is in this image?” Submit an image, and the service returns scores for relevant classifiers representing things such as objects , events and settings.

b) We feed TV stations audio and Radio stations audio to Speed to Text API

Watson Speech to Text can be used anywhere there is a need to bridge the gap between the spoken word and its written form. This easy-to-use service uses machine intelligence to combine information about grammar and language structure with knowledge of the c...