top of page
Screenshot 2023-03-20 at 12.19.31 PM.png

 “Jack the Bulldog'” is a chatbot built for Georgetown Undergraduate Admissions Office designed to answer  commonly asked questions about admissions.

“Jack the Bulldog'” is a chatbot built with Rasa for the Georgetown Undergraduate Admissions Office designed to answer commonly asked questions about admissions, the application process, and selective topics about the university. A persona, Jack the Bulldog, is incorporated into the system to make the chatbot more fun, engaging, and likable.

Duration

2 months

Deliverables

Chatbot UI

RASA dialogue system

My Role   

Conversation Design

Natural Language Processing

The Idea

Deciding which college to attend is a big decision, and it is best to know as much information about the school you are applying to or planning to attend. The Georgetown Undergraduate Admissions Office receives roughly 100 emails and 100 phone calls each day, with higher numbers as application deadlines approach. Considering that Georgetown’s student population is one of the most geographically diverse in the country, representing every state and over 130 countries, it may be difficult for all prospective students with questions from around the world to contact the office during working hours. Therefore, our chatbot was designed to automatically answer commonly asked questions about the undergraduate application and school so that people can ask questions anytime.

Data Collection

Data were manually extracted from the Georgetown University websites, including the undergraduate admissions page and pages that branched from the main university page. The information from these webpages was written into responses for specific user intents.

Intents:  

Topics of input that the user could say that have a matching response from the system. A total of 47 intents with an average of about 10 intent examples for each.

Screenshot 2023-09-21 at 11.45.36 PM.png
Screenshot 2023-09-21 at 11.45.27 PM.png

Examples of good mood intent and challenge bot intent

Design the dialogue flow

To organize our intents, we created 11 categories that were related to topics one would find on a Georgetown University webpage, specific to the application process or the university in general: transfer students, dates, application requirements, admission statistics, international students, financial aid, housing, student life, visits, contact information, or out of domain.

Screenshot 2023-09-22 at 12.00.08 AM.png

Click to enlarge

Screenshot 2023-09-22 at 12.06.43 AM.png

Train the system

Entities are specific pieces of information that Rasa could extract from a user’s message, and responses are all of the system’s outputs. The responses we created were specific to each user intent,

The majority of our dialogue management was incorporated in Stories. For each intent, we created a story that offered a response to the user’s question and then asked for confirmation if the question was answered. We incorporated checkpoints that helped the conversation flow in the proper direction, based on whether or not the user was satisfied with Jack’s response and the information that was presented.

Screenshot 2023-09-22 at 1.08.45 AM.png

Integrate the persona

Entities are specific pieces of information that Rasa could extract from a user’s message, and responses are all of the system’s outputs. The responses we created were specific to each user intent,

Screenshot 2023-09-22 at 1.17.02 AM.png

We chose Jack the Bulldog, Georgetown University’s official mascot, as our persona. The system embodies the persona mainly through the use of language.

Jack has a unique way of saying things and it is incorporated throughout the conversation. For example, he calls the admission office “Paw-ffice” and he apologizes by saying “I a-paw-logize.”

Screenshot 2023-09-22 at 1.20.42 AM.png

Jack tells a fun fact about Georgetown when he cannot answer a question, and he is willing to provide more fun facts if people ask for it.

Screenshot 2023-09-22 at 1.20.21 AM.png

We tried to show his emotions and attitudes in the conversation.

Screenshot 2023-09-22 at 1.57.34 AM.png

VUI Design

GU's mascot logo

Screenshot 2023-09-22 at 2.23.46 AM.png

Hamburger menu for restarting the session and Help option

Screenshot 2023-09-22 at 2.26.31 AM.png
Screenshot 2023-09-22 at 2.23.08 AM.png

Two of GU's official colours

Screenshot 2023-09-22 at 2.29.58 AM.png
Screenshot 2023-09-22 at 2.29.58 AM.png

Test Out the Design among Universities

To evaluate the system, both objective and subjective data were collected. A total of 16 participants were recruited to interact with the bot. Immediately following the interaction with the system, each participant was asked to complete a user experience survey.

Screenshot 2023-09-22 at 2.47.03 PM.png

Survey Responses on System Performance

Screenshot 2023-09-22 at 2.46.54 PM.png

Survey Responses on Persona

For the objective evaluation, we have two sets of data: one measures the overall success rate of the system (Table 1), and the other measures the task success rate (Table 2).

Screenshot 2023-09-22 at 3.15.04 PM.png

Out of 212 utterances, the system was able to respond correctly to 172, giving a success rate of 81.10%. The success rate decreased to 66.7% when only questions from users were considered.

bottom of page