PROJECT DESIGN PHASE-II

SOLUTION REQUIREMENTS
(FUNCTIONAL AND NON-
FUNCTIONAL)

Date 22nd September 2023

Team ID Team-592973
Project Name AIRLINE REVIEW CLASSIFICATION USING
MACHINE LEARNING
Maximum Marks 4 Marks

SOLUTION ARCHITECTURE:
FUNCTIONAL REQUIREMENTS:
The following are the key functional requirements for the automated airline review sentiment analysis
system which can be categorized into the following categories:

FR FUNCTIONAL REQUIREMENT(EPIC) SUB REQUIREMENT(STORY/SUB-TASK)

No.
FR-1. DATA COLLECTION • The system must collect data from various
platforms like- review websites, social
media, customer feedback form, etc.
FR-2. DATA PRE-PROCESSING • Text cleaning and preprocessing should
remove noises such as HTML tags, and
special characters
• If the data is in multiple languages, it
should include language detection and
translation capabilities.
• The system must extract information from
reviews and contexts.
FR-3. DATA STORAGE • The reprocessed data must be stored in a
structured database for efficient retravel
and analysis.
FR-4. SENTIMENT ANALYSIS • The system must classify the sentiment of
each review into various categories like
positive, neutral, negative etc.
• The system must utilize NLP (Natural
Language Processing (techniques and
machine learning models for sentiment
analysis.
FR-5. FEEDBACK CATEGORIZATION • The system must be able to categorize
reviews like in-flight experience, baggage
landing, customer service, etc.
• The system must use clustering or top
modelling algorithm for efficient
classification.
FR-6. USER AUTHENTICATION AND • Implement user authentication and
AUTHORIZATION authorization to control system access and
protect user data.
FR-7. WEB INTERFACE AND RESPONSIVENESS • Provide a user-friendly web interface for
airlines and staff to interact with the
system.
• Include dashboards for monitoring
sentiment analysis results and categorized
feedback.
• Enable airlines to respond to customer
feedback directly from the platform.
 FR-8. APIs • Develop APIs to allow integration with
other airline systems for accessing
sentiment analysis results and categorized
feedback.
FR-9. REPORTING AND INSIGHTS • Generate reports and visualizations that
provide insights into customer sentiment
and trends.
FR- NOTIFICATION SYSTEM • Implement a notification system to alert
10. airlines' staff or management about critical
feedback that requires immediate
attention.
• Notifications can be sent via email, SMS,
or in-app messages.
FR- SCALABILITY • The system should be designed to scale as
11. the volume of data and users grows,
potentially using cloud-based solutions.
FR- SECURITY • Ensure robust security measures to protect
12. user data and maintain data privacy.
• Encryption, access controls, and
compliance with data protection
regulations (e.g., GDPR) are necessary.
FR- CONTINUOUS TRANING AND • Periodically retrain the sentiment analysis
13. IMPROVEMENT models to adapt to evolving language and
feedback trends.
FR- DATABACK UP AND RECOVERY • Regularly backup the data and implement
14. disaster recovery plans to prevent data
loss.
FR- MONITORING AND ALERTS • Set up system monitoring to track
15. performance and health.
• Implement alerts for anomalies or issues.
FR- COMPILENCE • Ensure that the system complies with
16. aviation industry regulations and data
protection laws.
FR- INTEGRATION • Integrate the system with existing airline
17. management and customer relationship
systems for a holistic view of customer
feedback and operational performance.
FR- TESTING AND QUALITY ASSURANCE • Conduct rigorous testing, including unit
18. testing, integration testing, and user
acceptance testing to ensure system
reliability and accuracy.
FR- DOCUMENTATION • Create comprehensive documentation that
19. covers the architecture, data flows, and
system components for reference and
future maintenance.
FR- MAINTENANCE AND SUPPORT • Provide ongoing maintenance, support,
20. and updates to ensure the system's
reliability and accuracy.

NON- FUNCTIONAL REQUIREMENTS:

Non-functional requirements describe the qualities and characteristics that the automated airline
review sentiment analysis system should possess. These requirements focus on aspects like
performance, scalability, security, and usability.
The following are the key non-functional requirements for the automated airline review sentiment
analysis system which can be categorized into the following categories:

NFR NON-FUNCTIONAL SUB REQUIREMENT(STORY/SUB-TASK)

No. REQUIREMENT(EPIC)
NFR- USABILITY • User-Friendly Interface: The web interface
1. should be intuitive and user-friendly for
airline staff.
• Accessibility: Ensure that the system is
accessible to individuals with disabilities.
• Multilingual Support: If applicable,
support multiple languages for user
interfaces and customer feedback analysis.
NFR- SECURITY • Data Security: Ensure data encryption
2. during transmission and storage to protect
sensitive information.
• User Authentication: Implement secure
user authentication and authorization
mechanisms.
• Compliance: Adhere to data protection
regulations (e.g., GDPR) and aviation
industry security standards.
• Vulnerability Assessment: Regularly
perform security audits and vulnerability
assessments.
NFR- RELIABILITY • Fault Tolerance: The system should be
3. resilient to hardware or software failures,
minimizing service interruptions.
• Data Integrity: Ensure the accuracy and
integrity of data throughout the processing
pipeline.
• Backup and Recovery: Implement a robust
data backup and recovery strategy to
prevent data loss.
NFR- PERFORMANCE • Response Time: The system should
4. provide near real-time sentiment analysis
and categorization of reviews.
• Throughput: The system should handle a
high volume of concurrent requests and
data processing.
• Scalability: The architecture must support
horizontal scalability to accommodate
growing data and user loads.
• Availability: The system should have a
high availability rate, with minimal
downtime for maintenance and updates.
NFR- AVAILABILITY • High Availability: The system should aim
5. for a minimum of 99.9% availability,
ensuring it is accessible to users at almost
all times.
• Monitoring and Alerting: Utilize
monitoring tools to detect service
disruptions or anomalies and trigger
automatic alerts to the system
administrators.
• Disaster Recovery: Develop a
comprehensive disaster recovery plan that
includes offsite backups, data restoration
procedures, and clear guidelines for
system recovery in case of catastrophic
failures.
• Scheduled Downtime: Schedule
maintenance and updates during off-peak
hours to minimize disruption to users,
 with advanced notification of downtime.

NFR- SCALABILITY • Load Balancing: Implement load

6. balancers to evenly distribute incoming
traffic across multiple servers or instances
to ensure optimal system performance.
• Database Scalability: Ensure the database
system is capable of scaling horizontally
to manage growing data volumes and
concurrent requests.
• Caching: Implement caching mechanisms
to reduce the load on the database and
improve response times, particularly for
frequently accessed data.