IMPROVING RURAL

HEALTHCARE ACCESS IN
KENYA WITH
CONVERSATIONAL AI
PROBLEM STATEMENT
•Rural populations in Kenya have extremely limited access to
healthcare and health information
•Key challenges:
• Few health facilities, long distances
• Shortage of doctors and nurses
• Lack of awareness on health issues
•This leads to delayed diagnosis and care seeking, resulting in poor
health outcomes

© 2023
Falcon AI Chatbot
PROPOSED SOLUTION

HEALTH BUSINESS

Additionally, it aligns with For businesses, it streamlines

environmental and health operations, enhances

concerns by promoting competitiveness, and

sustainable pest control contributes to sustainability.

practices.

© 2023
Falcon AI Chatbot 3
 IMPORTANCE
1. Technology Advancements: With the rapid advancements in artificial intelligence and natural

language processing, AI chatbots have become more sophisticated, making it possible to offer

highly efficient and responsive services.

2. Customer Expectations: Customers want instant feedback and round-the-clock access to

information.

3. Environmental and Health Concerns: Increasing concerns about the environmental impact of pest

control methods and the safety of chemicals used to provide the need for more education and eco-

friendly alternatives, which chatbots can provide.

© 2023 Falcon AI Chatbot 4
 IMPORTANCE
4. Business Efficiency: For pest control companies, streamlining operations and improving

efficiency are essential to stay competitive in a crowded market. Chatbots can automate

routine tasks, freeing up human resources for more complex issues.

5. Data-Driven Decision-Making: The ability of AI chatbots to collect and analyze data provides

valuable insights into customer preferences, common pest problems, and market trends,

allowing businesses to adapt and make informed decisions.

© 2023
Falcon AI Chatbot 5
 STATEMENT OF THE PROBLEM/OPPORTUNITY

• The traditional pest control industry faces challenges related to meeting evolving customer

expectations for immediate and convenient support.

• Additionally, there’s a need for businesses to streamline operations, enhance

competitiveness, and contribute to sustainability.

• These challenges and opportunities necessitate the development of an AI-powered chatbot to

modernize the industry, improve customer experiences, and address health, safety, and

environmental concerns.
© 2023
Falcon AI Chatbot
 AREA OF RESEARCH
 The primary area of research centres on AI Chatbot development. This encompasses
advances in Natural Language Processing (NLP) for understanding and generating
human-like text responses and Machine Learning and AI models to enhance the
chatbot's learning capabilities.

 Research also includes data analytics for gaining insights from user interactions and
security measures to protect user data and ensure compliance with privacy
regulations.

© 2023
Falcon AI Chatbot
 TOOLS AND TECHNOLOGIES TO APPLY

Natural Language Processing (NLP) NLP models like BERT, GPT-3, to enable the

chatbot's understanding of and interaction with users in natural language.

© 2023
Falcon AI Chatbot
 TOOLS AND TECHNOLOGIES TO APPLY

Machine Learning and AI Models: Algorithms and models for continuous

improvement in chatbot responses, decision-making, and user experience.

© 2023
Falcon AI Chatbot
 TOOLS AND TECHNOLOGIES TO APPLY

Data Analytics and Reporting: Advanced analytics tools and dashboards to extract insights from

user interactions.Use Python libraries (e.g., Pandas and NumPy) for data collection and analysis.

© 2023
Falcon AI Chatbot
 TOOLS AND TECHNOLOGIES TO APPLY

 User-Centric Design: Figma Tool for the UI/UX design to ensure a user-friendly and intuitive interface.

 Web Development: HTML, CSS, JavaScript, React JS framework.

© 2023
Falcon AI Chatbot
© 2023
Falcon AI Chatbot 12
 MAIN OBJECTIVE

The primary objective of this project is to design, develop, and deploy a

web based AI-powered Pest Control chatbot that revolutionizes customer

experiences and operational efficiency in the pest control industry.

© 2023
Falcon AI Chatbot 13
SPECIFIC OBJECTIVES

1. To develop NLP capabilities that enable the chatbot to comprehend and respond to a

wide range of customer inquiries related to pest control, ensuring that it can handle

diverse user interactions effectively.

2. To implement a seamless connection between the chatbot and Falcon’s databases

containing information about common pests, their habits, and eco-friendly solutions, as

well as integration with scheduling systems for appointment bookings and updates.

© 2023
Falcon AI Chatbot
SPECIFIC OBJECTIVES (CONTINUED)
3. To develop and fine-tune machine learning models that enable the chatbot to
continuously improve its responses based on user interactions, thus ensuring its
accuracy and relevancy over time.

4. To perform data collection and analysis procedures to extract meaningful insights from
user interactions, trends in pest-related inquiries, and customer preferences, guiding
improvements in the chatbot's performance.

5. To create an intuitive, user-friendly interface for the chatbot, emphasizing ease of use
and responsiveness to enhance user satisfaction.

© 2023
Falcon AI Chatbot
 STATE OF THE ART
AUTHOR TOPIC ACHIEVMENT
OpenAI (2021) ChatGPT: A Large-Scale Generative Model for It demonstrated the capabilities of large-scale
Computer-Generated Conversations" generative models in creating human-like text.

Liang-Chih Yu, Hui-Lien Tung, and Yu-Hao Chu A Survey of Chatbot Implementation in Customer This survey provides insights into implementing
(2019): Service: What Works and What Doesn't chatbots in customer service and outlines best
practices.

Sowmya Vajjala, Bodhisattwa Majumder, and Anuj Practical Natural Language Processing: A This guidebook offers insights into building real-
Gupta (2020) Comprehensive Guide to Building Real-World NLP world NLP applications, including chatbots.
Systems

Matthew E. Peters et al. (2018) ELMo: Deep contextualised word representations" This paper introduced ELMo (Embeddings from
Language Models), which demonstrated the
effectiveness of contextual word embeddings in NLP
tasks.

Jacob Devlin et al. (2018): BERT: Pre-training of Deep Bidirectional This paper introduced BERT, a model that
Transformers for Language Understanding revolutionized how NLP tasks are approached by
pre-training on large text corpora.

© 2023
Falcon AI Chatbot
APPROACH/ METHODOLOGY

© 2023
Falcon AI Chatbot
APPROACH/ METHODOLOGY (CONTINUED)

© 2023
Falcon AI Chatbot
MACHINE LEARNING PIPELINE

INPUT DATA PREPARATION MODELING EVALUATION OUTPUT

IMPORT MODEL
REMOVE DUPLICATES STRUCTURE
DATA SELECTION CONVERSATION

REPLY

TEXT TO NUMERIC
EVALUATE ACCURACY

TRAIN THE MODEL

SPLIT DATA IMPROVE

(RETRAINING)

© 2023
Falcon AI Chatbot
 DATA

© 2023
Falcon AI Chatbot
DATA
1. Primary Data Sources:

 User Interactions: This includes chat logs, user queries, and chatbot responses.

 Appointment Data: Integration with scheduling and appointment systems.

 User Feedback: User feedback, including ratings, comments, and specific

suggestions.

© 2023
Falcon AI Chatbot
3.4. EVALUATION /TESTING /VALIDATION

1. Unit Testing: Individual components of the chatbot, such as NLP modules, machine learning

models, and data integration components.

2. Integration Testing: This includes testing data flow, API integrations, and system communication.

3. Functional Testing: Test cases will include common pest-related queries, appointment scheduling,

responses to user feedback, and integration with Falcon Fumigating Company's scheduling

systems.

4. User Acceptance Testing (UAT): Employees and selected customers will participate in UAT to

assess the chatbot's usability and its alignment with their specific needs.

5. Load and Performance Testing: Stress testing to determine its performance under high user load.
© 2023
Falcon AI Chatbot
3.4. METRICS FOR EVALUATION

Response User
Response
Time Satisfaction
Accuracy

System
Error Rate Conversion
Perfomance
Rate

© 2023
Falcon AI Chatbot
EXPECTED OUTCOMES AND DELIVERABLES:
 The primary outcome of the project will be the deployment of a functional AI-powered Pest Control
chatbot that is integrated with Falcon Fumigating Company's systems and is accessible to
customers.

1. Authentication and Security. – Entity recognition

2. User intent recognition. – Process user query

3. Analytics and Reporting.

4. User Experience Design.

5. Dialog Management.

© 2023
Falcon AI Chatbot
ETHICAL CONSIDERATIONS

Ethical Consideration Concern Measure

Privacy And Data Security Conversations may contain sensitive • Implement robust encryption for data
information, and data breaches could in transit and at rest. Anonymize user
lead to privacy violations. data for analysis.
• Ensure compliance with data
protection regulations, including
GDPR and HIPAA (if applicable).

Bias and Fairness AI chatbots can inadvertently perpetuate • Regularly audit and fine-tune the
biases in responses, which can lead to chatbot to minimize bias. Use
unfair treatment of users. diverse and representative training
data. Implement transparency and
fairness assessments to detect and
mitigate bias.

© 2023
Falcon AI Chatbot
ETHICAL CONSIDERATIONS
Ethical Consideration Concern Measure

1. Privacy And Data Security Conversations may contain sensitive • Implement robust encryption for data in transit
information, and data breaches could and at rest. Anonymize user data for analysis.
lead to privacy violations. • Ensure compliance with data protection
regulations, including GDPR and HIPAA (if
applicable).

2. Bias and Fairness AI chatbots can inadvertently • Regularly audit and fine-tuning.
perpetuate biases in responses, which • Use diverse and representative training data.
can lead to unfair treatment of users. • Implement transparency and fairness assessments

2. User Consent Obtaining informed consent for data • Clearly communicate data usage and seek user
collection and usage is essential. consent for data collection. Allow users to opt out
of data collection.

© 2023 Falcon AI Chatbot

CONCLUSION
- AI-powered Pest Control chatbot promises operational
excellence, sustainability, and innovation.

- Streamlines operational processes and promotes eco-

friendly pest control practices.

- Generates valuable data for Falcon Fumigating Company to

adapt to changing market dynamics and provide personalized
service.

- Multilingual support and round-the-clock accessibility break

down barriers to a diverse and global customer base.

- Has boundless potential to revolutionize service industries

through technology and set a standard for innovative,
customer-focused, and environmentally conscious solutions.
2023 Falcon AI Chatbot 27
REFERENCES
Jiang, H., Cheng, Y., Yang, J., & Gao, S. (2022). AI-powered chatbot communication with customers:
Dialogic interactions, satisfaction, engagement, and customer behavior. Computers in Human
Behavior, 134, 107329.

Adamopoulou, E., & Moussiades, L. (2020). Chatbots: History, technology, and applications. Machine
Learning with Applications, 2, 100006.

Adam, M., Wessel, M., & Benlian, A. (2021). AI-based chatbots in customer service and their effects
on user compliance. Electronic Markets, 31(2), 427-445.

Adler, R. F., Iacobelli, F., & Gutstein, Y. (2016). Are you convinced? A Wizard of Oz study to test
emotional vs. rational persuasion strategies in dialogues. Computers in Human Behavior, 57, 75-81.

© 2023
Falcon AI Chatbot
GANNT CHART

Literature Review Sys Analysis

SEM 1 JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC

SEM2 JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC

System Implementation
Deployment Deliver to client
Testing & Refinment

2/16/2024 HealthCare Chatbot 29

 THANK YOU

© 2023
Falcon AI Chatbot