Read ezine online
|by Jakub Kabourek
Issue 84 - December 2013
|The multi-screen revolution is being
driven by the consumer and is
developing quickly. It offers huge
potential and a bright future for operators
adopting OTT across multiple devices.
By delivering quality television services operators are
expanding their market potential and thereby significantly
increasing revenue and meeting the ever increasing
demands of their customer base.
But consumers dont just want to view content anytime,
anywhere, they want to be fed that content based on what
their friends are watching and they want to fi nd it quickly
and easily. To answer the needs of this ever growing market
nangu.TVs app available for download via the App Store or
Google Play provides users of its platform with a whole host
of features for multi-device and remote control use.
A key addition to the nangu.TV app is its Recommendation
Engine. There are three input elements: the metadata or
information about the content; the user behaviour; and the
priorities of the operators. Points one and three are clearly
defi ned, whereas point two, the user behaviour, is gathered
from several different places. To enable point two, the
Recommendation Engine tracks the selections made by
the user including: VoD channels; archived material; played
content; Time Shift viewed content; ratings; or content from
Between the input and the recommendation, the
processing engine works on the basis of four types of
algorithms: association rules; user behaviour; frequently
occurring series; and predictions based on genre, cast, etc.
All these algorithms work together to deliver the output.
Its important to set the weight of each algorithm to yield
the most accurate results for the given priorities that the
| In one way a Recommendation Engine is similar to a
universal framework that can be applied to any data.
Once data is input, the Recommendation Engine makes
a complex system of associations between the individual
elements. When an action, such as pressing play or
bookmarking is carried out, the Recommendation Engine
immediately highlights this as the focal point and reveals the
content that is in any way connected to it.
Additional clicks, such as clicking further detail, will
strengthen connections and generate new content.
Recommendations can show similar movies, genres,
date of production or other user preferences. The
Recommendation Engine can even go as far as to make
recommendations on similarity of scenes. Fingerprinting
allows for predictions to be made based on the visual effects; a feature that users may not even identify as a
criteria that affects their liking of a movie or series.
TV continues to be a source of entertainment and
comfort. Operators can provide this security with the
Recommendation Engine while increasing ARPU, a
small investment that leads to a larger margin. The more
comfortable the user feels that they can easily access
relevant content the more likely they are to remain loyal to
|Using the nangu.TV app its also possible for consumers to
control their TV using a tablet or Smartphone device. This
remote application enables simple browse capability with
on/off, play, pause, start again and volume adjustment.
Pop-up push notifi cations on the second screen alert the
user that content is starting and a single click plays it on
the TV. The search optimisation is highly advanced enabling
fi ltering based on chosen criteria. Users can search, record,
bookmark or play content on the TV freeing up the second
With the integration of apps on mobile, tablet and smart
TVs, watching OTT content across multi-screen devices
has never been easier. Users of the nangu.TV platform
simply purchase the app and access content on the
platform using the pin number provided by their operator.
| multi screen
| over the top
| Jakub Kabourek
Article Copyright tv-bay limited. All trademarks recognised.
Reproduction of the content strictly prohibited without written consent.