2015 Refugees Participatory Assessment in Jordan

Participatory assessment is a process of building partnerships with women and men of concern - of all ages and backgrounds - through systematic, structured dialogue on issues that they identify as important. In other words, people of concern to UNHCR describe their situation from their perspectives, voice their concerns, and mobilise their own efforts in solving problems within the community, thus taking control of their immediate future. Participatory assessment is an integral part of UNHCR programming instructions.

The UNHCR Tool for Participatory Assessment in Operations outlines 10 basic steps to ensure women, girls, boys, and men participate in analyzing protection problems together; in discussing capacities to face protection problems; and in finding solutions together. It offers a practical methodology for field teams, composed of UNHCR staff and partners, to operationalise protection and to support the implementation of a rights-based and community-based approach in their search for solutions to the protection problems of all people of concern.

During Participatory Asessment large amounts of qualitative information is gathered. Various subject matter expert are analysing this information in order to develop the report. In addition of those traditional approaches, the elements below provide the result of various Text Mining techniques

1- Terms Occurence & Word Cloud

Text mining aims at “...finding interesting regularities in large textual datasets...”. The first step is to define a corpus of all terms used within the collection of documents. It implies some processing stpes, like the elimination of ""stop words" (Stop-words are words that from non-linguistic view do not carry information).

2- Topic Modelling

A "topic model" is a type of statistical model for discovering the abstract "topics" (the hidden thematic structures) that occur in a collection of documents. Topic models capture the intuition of word occurrence and combination in a mathematical framework, which allows examining a set of documents and discovering, based on the statistics of the words in each, what the topics might be and what each document's balance of topics is. These algorithms allows for searching, browsing and summarizing large texts.

3- Correlation between topics and population group

Text mining aims at “...finding interesting regularities in large textual datasets...”. The first step is to define a corpus of all terms used within the collection of documents. It implies some processing stpes, like the elimination of ""stop words" (Stop-words are words that from non-linguistic view do not carry information).