Department of Information Engineering and Computer Science

Serious games are pedagogical tools used in many
domains and considered a viable solution to professional training.
Core elements for the construction of effective serious games are
the Non-Player Characters (NPCs) which, depending on the
quality of the Artificial Intelligence (AI), can make the training
scenarios more realistic and engaging. One of the main issues
with existing games concerns the limited possibility to edit their
content. These tools are usually delivered as closed products with
no, or minimal, customization features. The modification of the
virtual environment is strongly constrained and the behavior of
NPCs is fixed to a predefined set of actions. Therefore it becomes
hard for trainers to adapt the training according to different
requirements. In this paper we propose a solution for the
customization of AIs, aimed at allow non-technical users to edit
complex behavioral models through a new high-level approach.
The Actors’ Programming Environment (APE) has been
developed on the base of a user-centered design and a
preliminary evaluation was conducted with eight experts in
computer science and interaction design, serving as evaluators.
The results suggest that the approach is suitable for this context
and might prove to be a good basis for future developments.

This study, conducted by Dnet under a grant provided through USAID’s Mobile Solutions and Technical
Assistance (mSTAR) project, captures how customer acquisition agents for Dnet’s Aponjon Program are using
mobile phones. Aponjon is an mHealth initiative focused on maternal and child healthcare. A mix of
quantitative and qualitative methods were used for this study, which had a sample size of 178 agents, selected
from a total population of 1,029 agents. Of the 178 agents selected for the study, 86 are cash recipients
(referred to as “cash users”) and 92 are mobile money recipients (referred to as “MFS users”)1. Nine focus
group discussions and 29 key informant interviews were conducted with the agents. Excel and Google analyzer
were used to analyze the data and create graphs of the findings. Participants’ opinions were analyzed on
multiple questions using a Likert Scale. Key findings of the study are detailed below:
1. 84% of participants use a feature phone, and 5% access the internet using a mobile device. Interestingly,
while more cash users owned smartphones (21% vs. 12%), they were nine times less likely to use mobile
internet. Cash users were also significantly less likely than MFS users to use their phones for services
beyond just voice.
2. The majority of MFS users (88%) use mobile financial services to receive money, whereas cash users (71%)
collect money from their affiliated local office branch manager. 97% of MSF users felt that mobile financial
services are convenient, whereas 78% of cash users stated that hand-to-hand cash transactions are
convenient.
3. On average, it takes cash users approximately two hours (86 minutes of travel plus 33 minutes of wait
time) to withdraw their payment from an affiliated local office. In monetary terms, Dnet estimates the
time value to collect each payment equals a total of Tk. 51 (Tk. 37 for travel time and Tk. 14 for wait
time). Concerns about the time required to collect cash and the cost of travel were noted as
disadvantages of using cash by a significant portion of cash users (47% and 71%, respectively).
4. 62% of cash users (n=86) have knowledge of MFS. Interestingly, more than 20% of cash users expressed
positive feelings about receiving payments via MFS because they viewed it as safe, secure, convenient and
easy to operate. More than half of all participants learned about MFS through television advertisements
and/or conversations with colleagues.
5. While both types of participants noted that they save money, the percentage was much higher among cash
users (88%) compared to MFS users (40%). Awareness of interest rates offered by banks appears to be
the reason for the difference.
The study revealed that there is substantial potential to improve the capacity, knowledge and awareness of
MFS among cash users, particularly among those individuals who recognized some of the benefits of MFS.

Ontologies can formally describe the semantics of the medical domain in an unambiguous and machine processable form, acting as a conceptual interface between different applications that must interoperate. In this paper we present an ontology of cancer therapies originally developed to bridge the gap between an oncologic Electronic Patient Record (EPR) and a guideline-based decision support system. We show an application of the ontology complemented by rules to classify therapies recorded in the EPR. The results show how such an ontology can be used also to discover possible problems of data consistency in the EPR.