Page 431 - AI for Good Innovate for Impact

AI for Good Innovate for Impact

(continued)

Item Details
Key Aspects of a. Data mining: extracting anonymised data from UNHCR’s case management
Solution system. 4.4-Productivity
b. Exploratory data analysis:

• Process mining – map the process from the moment of the asylum applica-
tion until a decision is made to identify patterns.
• Big data analysis – analyse log data to understand bottlenecks (demo-
graphics and operational).
c. Machine learning (ML) and artificial intelligence (AI) to identify trends: train
the model using ML and AI algorithms to identify bottlenecks, identify
root-causes of delays, and predict the length of the RSD process.
d. Prescriptive analysis for efficiency gains: provide recommendations to avoid
delays, gaining efficiency in the RSD process under UNHCR’s mandate.

Technology Data mining, process mining, machine learning, artificial intelligence, linear
Keywords regression modelling, predictive analytics.
Data Availabil- The data source is private. Anonymised and aggregated data is extracted
ity from UNHCR’s case management system following privacy-by-design and
data-minimisation principles in compliance with UNHCR’s data protection and
privacy framework. Data is stored in a secure Microsoft Fabric environment
(subject to a framework agreement with UNHCR).
Metadata (Type Relational database, structured SQL tables, process logs, categorical data,
of Data) text data.

395