Salwan Public School, Mayur Vihar
Artificial Intelligence (843)
Project Documentation
Session: 2025-26
Project Name: ADD
School Name: Salwan Public School
Teacher name: Pooja Dhingra
Grade: 12th
Team Members:
Pradyumn Jha
Adarsh
Arpit Pal
Ananya Jayant
1. Prepare for the Project
1.1 Identify a local issue affecting your school or community that could
be solved using artificial intelligence (AI)
Senior secondary students encounter substantial academic demands due to competitive
examinations, evolving societal expectations, and rapid socio-technical change.
These demands occur alongside normal developmental challenges of adolescence, such as
identity formation, decision-making under uncertainty, and evolving peer and family
relationships. Many students are hesitant to seek formal counselling, not necessarily
because of the absence of institutional support but because of personal factors such as
unfamiliarity with psychological terminology, perceived stigma, limited time, lack of
resources, and uncertainty about what to expect from counselling.
Consequently, there is a clear need for accessible, private, and user-friendly tools that
enable students to record and reflect on their experiences, recognise recurring patterns in
mood and behaviour, and explore future options in a structured manner.
Artificial Intelligence can address this need by providing an automated, non-judgmental
reflective interface that augments existing school support systems, enhances students'
emotional literacy, and offers structured guidance for personal planning without replacing
professional services.
2. Team Roles
2.1 Team Members and Roles
2.1 Who is in your team and what are their roles?
Role Role Description Team Member Name
Project Leader Programming of the main Pradyumn
software and design,
assignment of tasks, review
of progress
Member FILL Adarsh
Member FILL Arpit
Member Documentation, Compilation Ananya
of data, records, and
minutes
2.2 Project Plan
Phase Task Date Expected Actual duration
duration
Preparation Coursework, 23/06/25 2-3 hrs 2-3 hrs
reading
Formation of Familiarising, 23/6/25 1-2 hrs 1.5 hrs
Team interaction with
peers
Defining the Background 24/6/25 4-5 hrs 6 hrs
Problem reading,
research
Understanding Identifying the 27/6/25, 28/6/25 7-8 hrs 7 hrs
Users
Meeting,
observing and
interacting
Approach Generating 30/6/25, 2/7/25 6 hrs 5.5 hrs
ideas, Voting for
most suitable
idea
Data Collection Analyzing Data 7/7/25 4 hrs 4.5 hrs
Requirements
Designing Developing AI 14/7/25, 24-25 hrs 28 hrs
Model model 16/7/25, 18/7/25
Prototype
Prototype Internal testing, 20/7/25, 21/7/25 3-4 hrs 3 hrs
testing multiple user
testing
Reflection on Reflecting with 22/7/25 1-2 hrs 1.5 hrs
the project Team members
Creating Pitch
video
2.3 Communication Plan
1. Did you meet physically, online or a mixture of each to communicate?
Ans: Mixture
2. How often did you come together to share your progress?
Ans: Weekly
3. What tools did you use for communication?
Ans: Google Drive, WhatsApp, Gmail.
2.4 Team Meeting Minutes
Meeting 1: Preparation & Team Formation
● Purpose: Establish project scope and form team roles
● Items Discussed: Review of capstone requirements, selection of theme (student
well-being), identification of preliminary responsibilities (research, testing,
documentation, coding).
● Things To Do: Compile background literature; confirm role allocations; schedule
follow-up meeting.
Meeting 2: Problem Definition & Understanding
● Purpose: Define the local issue and gather contextual information
● Items Discussed: Academic pressures in the local school, barriers to accessing
counselling, potential AI interventions, ethical considerations.
● Things To Do: Prepare interview questions; identify student volunteers for
observation.
Meeting 3: User Understanding (Observations & Interviews)
● Purpose: Conduct observations and short interviews with peers
● Items Discussed: Patterns of study habits, emotional vocabulary gaps, informal
coping strategies.
● Things To Do: Analyse interview responses.
Meeting 4: Idea Generation and Selection
● Purpose: Brainstorm AI solutions and select the most feasible concept
● Items Discussed: Emotional journal analyser, future pathway simulator, narrative
synthesis (weaver) and dream interpreter.
● Things To Do: Vote on integrated solution; document data requirements.
Meeting 5: Data Requirements & Collection Strategy
● Purpose: Finalise data and collection process
● Items Discussed: Types of textual entries required, anonymity and consent
procedures, storage and retention.
● Things To Do: Collect initial anonymised sample entries.
Meeting 6: Prototype Development (Module Review)
● Purpose: Review initial model outputs and interface mockups
● Items Discussed: Emotion scoring logic, navigator output format, integration points
between modules.
● Things To Do: Refine prompt templates; implement weighted average emotional
composition.
Meeting 7: Prototype Development (Model Refinement)
● Purpose: Debugging and model performance tuning
● Items Discussed: False positives in emotion detection, handling ambiguous
language, UX copy for safety disclaimers.
● Things To Do: Adjust model thresholds; prepare user-testing plan.
Meeting 8: Prototype Development (Final Build)
● Purpose: Final integration and deployment to test environment
● Items Discussed: Deployment checklist, test user accounts, basic logging.
● Things To Do: Deploy staging instance; inform test users.
Meeting 9: Prototype Testing
● Purpose: Internal testing and peer testing session
● Items Discussed: User comprehension, response clarity.
● Things To Do: Collect feedback; prioritise fixes.
Meeting 10: Prototype Testing (Follow-up)
● Purpose: Review testing feedback and assign refinement tasks
● Items Discussed: Clarification of technical language, additional journaling prompts.
● Things To Do: Implement fixes; prepare reflection notes.
Meeting 11: Reflection and Documentation
● Purpose: Final project reflection and preparation of submission
● Items Discussed: Lessons learned, scope for future work, division of final report
sections.
● Things To Do: Complete documentation; record pitch video.
Module-specific Minutes (one per module)
1. Mirror Module Review
● Purpose: Evaluate emotion-scoring algorithm and output phrasing
● Items Discussed: Emotion range (joy, anxiety, calm, sadness, anger), scoring scale
(0–10), weighted average methodology, sample outputs and safety wording.
2. Navigator Module Review
● Purpose: Assess probabilistic pathway generation and recommendation clarity
● Items Discussed: Career archetype mapping, confidence percentages.
3. Weaver Module Review
● Purpose: Validate narrative and discuss phrasing of module
● Items Discussed: Minimum entry requirement for weaving (at least three threads),
handling of repetitive content.
4. Dream Decoder Review
● Purpose: Review symbolic mapping and multi-lens interpretation
● Items Discussed: Freudian and Jungian lenses, cautionary language about
interpretation limits.
3. Problem Definition
3.1 Important Local Issues
● Academic stress among senior secondary students
● Difficulty expressing emotions and mental states
● Anxiety related to future careers and life choices
● Limited access to private, stigma-free emotional support
3.2 Issues That Matter Most
The issue that matters most is students’ inability to understand and process their emotional
state while simultaneously managing academic and future-related pressure. This affects
concentration, motivation, and overall well-being.
3.3 Focused Issue
The project focuses on providing students with a reflective AI-based tool to improve
emotional awareness and structured self-reflection.
3.4 Team’s Problem Statement
Students in high-pressure academic environments lack accessible tools to reflect on their
emotions, recognize mental patterns, and explore future possibilities, leading to stress,
confusion, and reduced well-being.
4. Understand Your Users
4.1 User Identification
The primary users are students experiencing a pressurising environment, leading to stress
and mental health issues.
4.2 Observations
Students often experience anxiety, overthinking, and emotional fatigue. Many find it difficult
to verbalise their feelings and hesitate to approach counsellors.
4.3 Interview Questions and Responses
Q1: Do you often feel stressed about academics or the future?
Common Response: Yes.
Q2: Do you find it easy to express your emotions?
Common Response: No, it is difficult to explain how I feel.
Q3: Would you use a private AI-based reflection tool?
Common Response: Yes, if it is non-judgmental and easy to use.
4.4 Empathy Map
Say Think
“I’m stressed” “Am I doing enough?”
“I don’t know what I want” “What if I fail?”
Do Feel
Anxious, tired
Avoid Proper discussion Confused
Indulging in unhealthy, non-sustainable
coping mechanisms
4.5 Current Steps and Difficulties
Students either suppress emotions or discuss them informally. The lack of structured
reflection tools makes emotional understanding difficult.
4.6 Refined Problem Statement
There is a need for a structured, private, and accessible AI tool that enables students to
reflect on emotions and explore future choices without stigma.
5. Brainstorming
5.1 AI Ideas
1. AI-based emotional journal analyser
2. AI future pathway explorer
3. AI dream and symbolism reflection tool
5.2 Priority Table
Idea User Value Ease of Implementation
Emotional Analyzer High Medium
Future Explorer High Medium
Dream Decoder Medium Medium
5.3 Selected AI Solution
The best solution is an integrated reflective AI system combining emotional analysis,
future exploration, and symbolic reflection.
6. Design Your Solution
6.1 User Steps
1. The AI solution is designed as a modular system where students may use one or
more components independently. Each module follows a simple
input–analysis–output structure while serving a distinct purpose.
Module 1: The Mirror – Reflective Intelligence Engine
1. The user opens the Mirror module on the web interface.
2. A prompt box appears asking:
“What’s the texture of your
mind right now?”
3. The user writes a short or long
journal entry in free text form.
4. An optional Deep Analysis
Mode can be turned on or off
depending on the level of detail
the user wants.
5. After submission, the AI
analyses the linguistic intensity
of the text.
6. The system evaluates five
emotional dimensions: Joy,
Anxiety, Calm, Sadness, and
Anger.
7. Each emotion is assigned a
numerical score on a scale of
0–10.
8. The system stores the
emotional scores from each
entry.
9. A visual chart is generated that
displays the weighted
average of emotional scores
derived from the user’s last
ten journal entries.
10.The user receives a short
reflective summary explaining
dominant emotions and
noticeable shifts over time.
11.This allows the user to observe
emotional patterns without
judgment or diagnosis
Module 2: Navigator – Probabilistic Future Simulator
1. Three input fields are displayed:
a. Skills
b. Interests
c. Personality traits
2. The user fills in these fields based
on self-perception
3. The action button labelled
“Generate Future” is clicked.
4. The AI analyses overlaps between
skills, interests, and personality
traits.
5. Using probabilistic reasoning, the
system generates exploratory
future pathways.
6. The output presents possible roles
or directions with brief explanations.
7. These outputs are not predictions
but structured possibilities meant to
encourage reflection.
Module 3: The Weaver – Literary Pattern Recognition
1. The Weaver module runs
automatically once data exists in
the Mirror and Navigator
modules.
2. No direct user input is required.
3. The AI combines emotional
trends from journal entries and
future-oriented data from
Navigator.
4. A narrative “chapter” is
generated describing a current
life phase or theme.
5. The output is written in
structured, readable language
rather than poetic abstraction.
6. The user reads the narrative to
understand how thoughts,
emotions, and aspirations
connect over time.
Module 4: Dream Decoder – Subconscious Translation Interface
1. The user opens the Dream Decoder
module.
2. A prompt box asks:
“Describe your dream in vivid detail
(include characters, settings,
emotions, and any unusual
elements).”
3. A supporting line reads: “Describe as
much as you remember.”
4. The user clicks the “Decode Dream”
button.
5. The AI identifies symbolic elements,
emotional undertones, and recurring
motifs.
6. The output presents interpretations
using psychological and symbolic
frameworks.
7. This helps users reflect on
subconscious themes without treating
dreams as literal predictions.
7. Identify and Collect Data
7.1 Data Type
● Textual journal entries
● Dream descriptions
● User-provided skill and interest data
7.2 AI Tools Used
● ADD
7.3 Outputs and Actions
The AI generates emotional scores, reflective narratives, and future pathways. Users can
use insights for self-reflection and decision-making.
8. Testing
8.1 Test Users
Classmates and peers from the same academic environment.
8.2 Observations
Users found the tool engaging and helpful for reflection. Many reported increased emotional
clarity.
8.3 User Feedback Grid
Likes Dislikes Suggestions
Easy to use Needs more entries Add history view, Option to delete
entries
9. Team Collaboration
9.1 Collaboration Description
9.1 How did you actively work with others in your team?
Ans: We discussed ideas and gathered their suggestions while designing the prototype.
We communicated regularly with teachers, who acted as stakeholders, to get guidance and
ensure the project stayed aligned with the learning objectives.
We listened to user feedback during testing and worked collaboratively to improve the
prototype based on their inputs.
During group discussions, we shared our progress, asked questions, and offered help when
needed.
10. Individual Learning Reflection
10.1 Team Reflections
The project highlighted the emotional challenges faced by students and demonstrated how
AI can be used responsibly to support self-reflection.
ADD YOUR OWN
Pradyumn:
Adarsh:
Arpit:
Ananya:
11. Links
Project URL:
[Link]
YouTube Video Link: