Qatalog 1

Qatalog

Query, Assign, Tag, and Analyse

Qatalog is a multi-faceted tool for accessing and analyzing numerous data sources including PDF documents, radio, and historical social media data related to specific topics, themes, and discussions.

About Qatalog

Qatalog is an AI-powered tool that uses speech recognition technology developed by UN Global Pulse to ‘listen’ to public radio talk shows and automate the detection of words spoken during those shows. It also pulls in public Twitter streams building on one of several partnerships established by UN Global Pulse with private sector data providers on behalf of the UN system. Lastly, it allows users to upload PDF documents for analysis. At the end of 2019,  Qatalog offered users the option to choose from 39 different languages to analyse.

The name QATAlog describes the main pipeline for data analysis: Query, Assign, Tag, and Analyse. The tool allows users to extract useful information from sources of big data and analyse it for topics of interest using a combination of optimized manual annotations techniques and automatic helpers that include translation, geolocation, and machine learning-driven text classification. Users can visualise the volumes of annotated content over time and space, and can download the raw data for further analysis.

See Qatalog in Action

Qatalog - Query

The Qatalog Pipeline

Q

Query

Retrieve data related to a topic of interest.

A

Assign & Agree

Collaborate on a taxonomy and assign workflows.

T

Tag & Train

Train the machine learning model.

A

Analyse

Run the analysis and visualise the results.

Query

Assign & Agree

Tag & Train

Analyze

Qatalog has the potential to be adapted and adopted in multiple investigative scenarios. The fact that it is designed with and for UN colleagues makes it easy to use beyond in-depth and complicated machine classifiers, text analytics, and APIs.

We’re beta-testing the tool together with UN Entities during 2020 and hope to open-source parts of it at a later stage. Its modular and collaborative structure allows users to recommend new functions for specific needs and collaborate on further updates to the platform at large.

We hope the use of this tool will increase efficiency reducing the time needed to complete certain analyses that provide critical operational information. A platform like Qatalog can also mobilize research agendas from the academic community where AI experts contribute new models that can be imported into the tool.

We foresee a future where AI systems with humans-in-the-loop provide accessible, timely, inexpensive and relevant discussion-based insights. The key is to design and scale collaborative intelligence systems with user feedback. We hope that eventually, the extended use of Qatalog will allow us to create a global public good consisting of a library of AI models that are adapted to different analysis scenarios for development, humanitarian action, and peace & security.

Scroll to Top