JUICE: Innovative search engine for journalists

– JUICE is a digital prototype to support creative thinking by journalists during the early development of news stories, says professor Neil Maiden. Neil Maiden is one of the team members behind JUICE, and they have created an innovative search engine for journalists.
Neil Maiden is professor of Digital Creativity in the Faculty of Management at the Cass Business at the City University London. His current funded and doctoral research is in 3 main disciplines: digital creativity support for professional work, specifying and designing complex digital systems, and digital support to improve the care of older people with dementia.
Publisert: 13. juni 2016

Intelligent software for journalists

JUICE is implemented as a simple Add-on Sidebar and Dialog Box in Google Docs. – JUICE presents the journalist with creative clues and ideas about new angles for a news story that they are currently writing in Google Docs, says Professor Neil Maiden.

– The current version of JUICE searches for news information from Twitter and from 477 different news information websites. The software also retrieves information from Wikipedia to validate and structure the results of its creative searches.

JUICE emerged from a user-centered design process, and is currently designed to be used by general reporters who investigate a wide range of stories in different topics.

Four key software components

1) ”The interactive Sidebar” and ”Dialog box” deliver retrieved and structured information to the journalist to stimulate creative thinking within the Google Docs editor.

2) ”The JUICE creative search service” manipulates text from the Google Docs editor to generate search queries that from each input to the 6 implemented creative search strategies.

3) ”The separate Juicer prototype news aggregation service”, available from the BBC, searches over 470 primarily English news sources using creative search queries generated by JUICE.

4) The Twitter API, which searches for and retrieves more recent Twitter content also using the same creative search queries generated by JUICE.

Maiden points out that JUICE also invokes software services to retrieve information from other online sources such as Wikipedia, as a part of the creative search service.

Sophisticated search algorithm

The search algorithm combines advanced natural language processing, information retrieval and recommender system algorithms to mimic the expert creative strategies of experienced journalists. – The core algorithm was developed from extensive previous research at City University London on creative search algorithms for other domains such as design and healthcare. The algorithm was specialised using knowledge from domain experts in journalism, says Maiden.

Neil Maiden elaborates that the JUICE team has experimented with automatic language translation services such as Google Translate, and have found it to be sufficiently accurate to enable cross-language translation. For example, to discover relevant Norwegian language news items in response to English-language stories and inputs.

Interviews revealed 6 strategies

In order to build JUICE the team needed to collect requirements from experienced journalists. – The JUICE team ran semi-structured interviews with 6 experienced journalists, and 2 leading digital experts in journalism to elicit strategies that these journalists use to explore and write creative angles on news stories. They found 6 principles to design JUICE:

(I) Individuals: creatively explore different human angles in a news story based on the different people and roles associated from its background.

(II) Casual: creatively explore the background events that underpin a story to discover a new angle to the story from its background.

(III) Quirky: creatively explore unusual or comical information about a story as a means of using wit to report serious news.

(IV) Quantifiable elements: creatively explore numerical and quantified information about a news story in order to back it up with evidence in new and useful ways.

(V) Ramifications: creatively explore information about the consequences of events and actions related to a news story, to generate new and useful stories about future consequences.

(VI) Data visualizations: creatively explore different data sets and visualizations to generate new and useful ways about how to display information in a news story to encourage human insight.

Facts about JUICE

JUICE is a digital prototype to support creative thinking by journalists during the early development of news stories.

The algorithm combines advanced natural language processing, information retrieval, and recommender system algorithms.

Watch a demo of JUICE on YouTube. Click here to watch

Neil Maiden

The JUICE-team: Neil Maiden, George Brock, Konstantinos Zachos, and Jamie Lockerbie.

University of Bergen collaborates with the JUICE-project through an innovation action EU-proposal. The Bergen participants are professor Lars Nyre, professor Astrid Gynnild, and professor Ole Johan Mjøs.

Need for efficiency

Maiden explains that speed was identified to be an important factor in news journalism, and journalists needs to work quickly and needs to maximise the support of available at minimum effort.

–To build a first prototype, our team investigated how to design and implement the digital support for creativity within one such daily work tool, Google Docs, to support journalist creativity without the need to use other tools.

Maiden reveales that by the end of 2016 the team will develop a full working version of JUICE that supports all 6 creatrive search and support strategies.


Publisert: 13. juni 2016
linkedin facebook pinterest youtube rss twitter instagram facebook-blank rss-blank linkedin-blank pinterest youtube twitter instagram