3.4 Special Technical Documents

This lesson covers four formal academic / professional documents that follow specialised conventions:

1. Project Synopsis
2. Scientific Article / Research Paper
3. Dissertation / Thesis
4. Proposal

---

1. Project Synopsis

A project synopsis is a brief outline of a proposed project submitted to a supervisor / guide for approval before the actual work begins. Typical length: 3-10 pages.

Purpose

• Convince the supervisor that the project is worth doing
• Demonstrate understanding of the problem and approach
• Outline the scope, methodology, and timeline
• Identify resources needed
• Receive approval / feedback before starting

Structure

Sample Project Synopsis (Final Year)

                         PROJECT SYNOPSIS

  Title:           Smart Library Management System using AI-Based
                   Book Recommendation
  Student:         Rohit Jangra (Roll: CA2026-101)
  Guide:           Dr. Anjali Verma
  Department:      Computer Applications
  Date:            28 May 2026

  ─────────────────────────────────────────────────────────────────

  1. INTRODUCTION

  Most college library systems today are basic — they record issues
  and returns but cannot help students discover relevant books. As a
  result, students rely on word-of-mouth or random browsing.

  2. PROBLEM STATEMENT

  Existing library systems lack intelligent discovery, leading to
  under-utilised collections (only ~30% of books circulate annually
  in our college library) and student frustration when looking for
  relevant material.

  3. OBJECTIVES

   a. Build a web-based library management system with standard
      issue / return / fine / search features.
   b. Integrate an AI-based recommendation engine that suggests
      books based on a student's history and similar students'
      borrowing patterns.
   c. Provide a librarian dashboard for collection analysis.
   d. Achieve > 80% user-satisfaction rating in pilot tests.

  4. LITERATURE REVIEW

  Reviewed five existing open-source library systems (Koha, Senayan,
  Evergreen, OpenBiblio, NewGenLib). All provide standard issue /
  return but none ship with recommendation. Three commercial systems
  (Sirsi, Innovative, Ex Libris) have basic recommendation but cost
  beyond a college's budget.

  5. METHODOLOGY

   - Tech stack: React (frontend), Node.js + Express (backend),
     PostgreSQL (data), Python + scikit-learn (recommender)
   - Recommender: collaborative filtering on borrowing history
   - Methodology: Agile, 8 two-week sprints

  6. EXPECTED OUTCOMES

   - Deployable web application
   - Working AI recommendation engine
   - User documentation
   - Final report

  7. TIMELINE

   Sprint 1-2:  Requirements + Database design
   Sprint 3-4:  Backend APIs
   Sprint 5-6:  Frontend + Integration
   Sprint 7:    Recommender engine
   Sprint 8:    Testing + Deployment

  Total duration: 16 weeks (Jun-Sept 2026)

  8. REFERENCES

  [Standard reference list]

---

2. Scientific Article / Research Paper

A research paper is a formal academic document that reports original research to the scientific or technical community. Published in journals or conference proceedings.

Purpose

• Document and share new findings
• Build on / challenge existing knowledge
• Establish academic credibility
• Allow others to reproduce and verify

Structure — IMRAD format

The dominant format for scientific papers globally:

Plus:

• Title
• Authors and affiliations
• Abstract (~250 words)
• Keywords (5-7)
• Conclusion (often combined with Discussion)
• References (in journal-specific style)
• Acknowledgements (funding, advisors)

Sample paper structure

TITLE: A Lightweight Recommendation Engine for College Library Systems

AUTHORS:
  Rohit Jangra (rohit@example.edu)
  Dr Anjali Verma (anjali@example.edu)
  Department of Computer Applications, XYZ College

ABSTRACT (200-250 words):
  Brief summary covering:
   - Purpose / problem
   - Method
   - Key findings
   - Main conclusion

KEYWORDS: library systems, recommendation engine, collaborative
filtering, education technology, computer applications

1. INTRODUCTION
   1.1 Background
   1.2 Problem statement
   1.3 Contribution of this paper
   1.4 Paper organisation

2. RELATED WORK
   2.1 Existing library management systems
   2.2 Recommendation algorithms in commercial systems

3. METHODOLOGY
   3.1 System architecture
   3.2 Dataset
   3.3 Algorithm — Item-based collaborative filtering
   3.4 Evaluation metrics

4. RESULTS
   4.1 Experimental setup
   4.2 Comparison with baseline
   4.3 User satisfaction (pilot study, n=50)

5. DISCUSSION
   5.1 Interpretation
   5.2 Limitations
   5.3 Future work

6. CONCLUSION

REFERENCES (IEEE or APA style, 20-40 references typically)

APPENDICES

Conventions for research papers

Citation styles (most common)

---

3. Dissertation / Thesis

A dissertation (or thesis) is a long, detailed, original research document submitted for a degree (Master's, MPhil, PhD).

Length

Dissertation vs Research Paper

Standard Dissertation Structure

   FRONT MATTER:
   - Cover and title pages
   - Declaration by candidate
   - Certificate by guide
   - Acknowledgements
   - Abstract
   - Table of contents
   - List of figures, tables, abbreviations

   MAIN BODY:
   Chapter 1 - Introduction
       1.1 Background
       1.2 Problem statement
       1.3 Objectives
       1.4 Scope
       1.5 Thesis organisation

   Chapter 2 - Literature Review
       2.1 Theoretical foundations
       2.2 Prior work
       2.3 Research gap

   Chapter 3 - Research Methodology
       3.1 Approach
       3.2 Data sources
       3.3 Tools / techniques

   Chapter 4 - Design / Implementation
       4.1 System design
       4.2 Implementation details

   Chapter 5 - Results and Analysis
       5.1 Experiments
       5.2 Findings

   Chapter 6 - Conclusion and Future Work
       6.1 Summary of contributions
       6.2 Limitations
       6.3 Future directions

   BACK MATTER:
   - References (extensive)
   - Appendices (code, data, surveys)
   - Index (for PhD)
   - List of publications (for PhD)

Dissertation writing — tips

1. Start early — most students under-estimate by 2-3x
2. Set chapter-wise deadlines — work backwards from submission date
3. Choose a focused topic — too broad is fatal
4. Keep a research diary — daily notes save weeks of rework
5. Manage references with software — Zotero, Mendeley, EndNote
6. Draft → revise → revise → revise — first draft is always rough
7. Get feedback iteratively — share each chapter with guide
8. Build the bibliography from day one
9. Plan the defence — prepare for tough questions
10. Mind the formatting — most universities have strict guidelines

The Viva (oral defence)

For dissertation submission, you usually face an oral examination:

• 15-30 minute presentation
• Questions from internal and external examiners (30-60 minutes)
• Prepare for: motivation, methodology, limitations, future work
• Have answers ready for "What is the contribution of your work?"

---

4. Proposal Writing

A proposal is a persuasive document suggesting a specific course of action — for a project, research, funding, contract, sale, internal initiative — and asking for approval or resources.

Purpose

• Persuade decision-makers to approve and fund the proposed work
• Demonstrate understanding of the problem
• Show why the proposer is the right entity to do it
• Outline what will be delivered, when, at what cost

Types of proposals

Characteristics of a good proposal

Standard Structure of a Business / Project Proposal

   1. COVER PAGE
        - Title, submitter, date, recipient

   2. EXECUTIVE SUMMARY
        - One-page overview

   3. PROBLEM / OPPORTUNITY STATEMENT
        - The need being addressed

   4. PROPOSED SOLUTION
        - What you suggest to do

   5. METHODOLOGY / APPROACH
        - How you will do it

   6. DELIVERABLES
        - What will be produced

   7. TIMELINE / SCHEDULE
        - Gantt chart, milestones

   8. TEAM / QUALIFICATIONS
        - Why your team is right for this

   9. BUDGET / PRICING
        - Detailed costs

  10. RISK AND MITIGATION
        - Challenges + how you'll handle them

  11. EVALUATION CRITERIA
        - How success will be measured

  12. APPENDICES
        - CVs, case studies, references

Sample Proposal Snippet (Executive Summary)

EXECUTIVE SUMMARY

ABC Technologies is pleased to submit this proposal to XYZ Bank for
the design, development, and deployment of a mobile-first customer-
service chatbot.

Current state: XYZ Bank's call centre handles 200,000 customer
queries per month, with average wait time of 7 minutes and 25% of
queries unresolved on the first call.

Proposed solution: AI-powered chatbot handling 70% of queries
without human intervention, escalating the remaining 30% to agents
with full context.

Expected outcomes:
   - 50% reduction in call-centre volume in 6 months
   - Average response time: < 5 seconds (vs current 7 minutes)
   - First-contact resolution: > 85% (vs current 75%)
   - ROI: payback in 14 months; ongoing savings of ₹3.2 cr / year

Investment: ₹85 lakh (one-time) + ₹40,000 / month (maintenance)
Timeline: 5 months from approval to production
Team: 8 engineers, led by a 12-year banking-domain consultant

We have delivered similar solutions for two other Indian banks
(case studies in Appendix B). We are confident this initiative
will significantly improve XYZ Bank's customer experience while
delivering measurable cost savings.

We look forward to discussing this proposal at your earliest
convenience.

---

Proposal vs Report — exam table

---

Common Mistakes in Special Documents

---

Study deep

1. Special documents have stricter conventions than general writing. Synopsis, paper, dissertation, proposal — each has expected structure. Deviating from the convention signals inexperience.

1. The abstract / executive summary is the document. For papers, the abstract decides who reads further. For proposals, the executive summary decides who approves. Write it last, but treat it as the most important section.

1. Reference management is a real skill. Zotero / Mendeley / EndNote saves dozens of hours during writing. Learn one early in your academic career.

1. Originality is examined twice. Most universities run plagiarism checks (Turnitin, iThenticate). Modern AI-detection is increasingly active too. Original work is non-negotiable.

1. Proposals win on understanding, not promises. The bidder who clearly understands the client's problem usually wins — even if their solution isn't the fanciest.

Key Terms — Lesson 3.4

The terms below cover the four "special documents" — synopsis, research paper, dissertation, proposal — and the surrounding academic and business vocabulary.

Project Synopsis — A short summary document (5–10 pages) submitted at the start of a project — typically a final-year academic project — describing what will be done, why, how, by when. Approved by a guide / examiner before the project work begins. Sections: title, introduction, problem statement, objectives, literature review, scope, methodology, tools, expected outcomes, timeline, references.

Problem Statement — A clear articulation of the issue the project aims to solve — what is wrong with the current state, why it matters, who is affected, what would be different if it were fixed. The single most important section: a vague problem statement guarantees a wandering project.

Objectives (SMART) — The specific, measurable, achievable, relevant, time-bound goals the project will accomplish. "Build an admission portal that processes 1,000 applications per day with <2-second response time, by 31 March" is SMART; "Build a great admission portal" is not.

Literature Review — A survey of existing work related to the project's topic — published papers, books, prior projects, industry reports. Demonstrates the writer has done their homework, identifies gaps the project will fill, and establishes credibility. Required in synopses, dissertations, and research papers.

Scope and Limitations — A pair of sections in any synopsis or proposal. Scope states what the project will cover. Limitations states what it will not cover (and why) — out-of-scope features, constraints, assumptions. Setting expectations early prevents downstream conflicts.

Methodology — The planned approach for accomplishing the project — choice of programming language, framework, design pattern, data sources, evaluation criteria. The "how" of the project. In research, this section is especially critical for replicability.

Tools and Technologies — A specific section listing the tools, languages, libraries, hardware, and platforms the project will use. Lets the reader judge the technical viability and required skills.

Expected Outcomes / Deliverables — A list of what the project will produce — software artefacts, datasets, reports, presentations. Distinguishes the project's tangible outputs from its broader goals.

Project Timeline / Gantt Chart — A schedule of what will be done by when, often visualised as a Gantt chart with bars for each task. Lets the guide / committee judge whether the timeline is realistic.

Research Paper — A formal academic document reporting original research findings in a structured format. Published in peer-reviewed journals or conference proceedings. Follows the IMRaD structure (Introduction, Methods, Results, and Discussion) plus title, authors, abstract, keywords, conclusion, references.

IMRaD — The standard scientific-paper structure: Introduction (context, problem, hypothesis), Methods (what we did), Results (what we found), and Discussion (what it means). Standard in life sciences, social sciences, and increasingly in computer science conferences.

Title and Authors — The opening of a paper. Title should be specific and informative ("A Comparative Evaluation of Recommendation Algorithms for Indian E-Commerce" — not "Recommendation Systems"). Authors are listed with affiliations; the first author is conventionally the primary contributor.

Abstract (Academic) — A 150–250 word standalone summary of the paper — purpose, methods, key findings, main conclusion. Indexed in databases (Scopus, Google Scholar); often the only part many readers will see. Written last but placed at the front.

Keywords — A list of 5–10 search terms describing the paper's subject. Lets the paper be found in academic databases. Choose keywords that real searchers would use — be specific (e.g., "collaborative filtering" not just "AI").

Hypothesis — A specific, testable prediction that the research aims to confirm or refute. In computer-science research, hypotheses are often framed as comparative claims ("Algorithm A is more accurate than Algorithm B on dataset X").

Peer Review — The process by which a research paper is evaluated by independent experts in the same field before publication. Reviewers judge originality, methodology, soundness, and significance. Most reputable journals and conferences require peer review.

Citation / Reference — Acknowledging a source you have used or built upon. Citations appear in-text ((Smith, 2023)) and a full bibliographic entry appears in the references section. Citation lets readers verify claims and explore further.

Reference Management Software — Tools like Zotero, Mendeley, EndNote, Citavi that organise references, generate citations in any style, and integrate with word processors. Saves dozens of hours during writing; essential for any academic work.

Dissertation / Thesis — A long, original research document submitted as part of a degree requirement — Master's dissertation, PhD thesis. Far longer than a paper (typically 80–300 pages), more comprehensive, and original work in the field. Includes detailed literature review, methodology, findings, discussion, and contribution-to-knowledge claims.

Thesis Defence / Viva Voce — The oral examination at which a dissertation candidate defends their work before a committee. The committee tests understanding, methodology, and contribution. A make-or-break stage of graduate degrees.

Proposal — A persuasive document suggesting a specific action — typically a business proposal pitching a product/service to a client, a research proposal seeking funding, a project proposal seeking approval, a policy proposal seeking adoption.

Proposal Structure — Common sections: executive summary, problem statement, proposed solution, deliverables, timeline, methodology/approach, team and qualifications, pricing/budget, references and case studies, terms and conditions, appendices. The proposal must answer: what problem are we solving, how, by when, at what cost, why us?

Solicited vs Unsolicited Proposal — Two contexts. Solicited proposals are submitted in response to a Request for Proposal (RFP) — terms and evaluation criteria are predefined. Unsolicited proposals are proactive — the seller approaches the buyer without invitation. Solicited proposals win more often; unsolicited ones win bigger.

Request for Proposal (RFP) — A formal document issued by a buyer inviting vendors to submit proposals for a specific need. Specifies the scope, evaluation criteria, response deadline, and submission format. Common in government procurement, large enterprises, and any structured-bidding process.

Pitch / Pitch Deck — A short, slide-based proposal usually presented orally — typically 10–15 slides for a startup investor pitch. Covers problem, solution, market, team, traction, financials, ask. Highly persuasive; modern equivalent of the verbal sales pitch.

Executive Summary (in Proposal) — A 1–2 page standalone summary of the proposal — problem, solution, expected outcomes, investment, timeline. Often the only part the senior decision-maker reads. Written last, placed at the front, treated as the most important section.

Call-to-Action (CTA) — The specific next step the proposal asks the reader to take — "Sign and return by 30 June", "Schedule a meeting", "Approve the budget". Without a clear CTA, even a strong proposal stalls.

ROI (Return on Investment) — A financial metric showing the return generated for each rupee invested. Often included in business proposals to justify the spend: "₹85 lakh investment delivers ₹3.2 crore annual savings — payback in 14 months." Concrete ROI dramatically improves proposal acceptance.

Case Study / Testimonial — Evidence of past similar work — a brief description of a previous client, the problem solved, the result. Reduces the buyer's perceived risk. Often placed in an appendix or referenced throughout.

Boilerplate — Standard text reused across multiple proposals — company overview, team bios, methodology descriptions, terms and conditions. Time-saver if managed well; embarrassment if a wrong client name remains in a copy-paste.

---

Common exam question: "What is a project synopsis? Explain its elements." — Define; list 10-11 sections (title, intro, problem, objectives, lit review, scope, methodology, tools, outcomes, timeline, references).

Common exam question: "Explain the structure of a scientific / research paper using IMRAD format." — Define IMRAD; full structure (title, authors, abstract, keywords, intro, methods, results, discussion, conclusion, references).

Common exam question: "What is a proposal? Explain its characteristics and structure." — Define; characteristics (persuasive, customer-focused, specific, evidence-backed, etc.); 12-section structure.

Common exam question: "Differentiate proposal and report." — Tabulate 5-6 differences (purpose, time orientation, persuasive vs informative, audience, examples).

Self-check

Recall the four special documents and their conventions — answer, then check.

1. What is a project synopsis, and when is it submitted? (a brief 3-10 page outline of a proposed project, submitted to a supervisor/guide for approval before the work begins)
2. Expand IMRAD. (Introduction, Methods, Results, and Discussion)
3. What is the typical page length of a PhD thesis? (200-400+ pages)
4. What is the difference between a solicited and an unsolicited proposal? (solicited is submitted in response to an RFP / Request for Proposal; unsolicited is pitched without being asked)
5. In the proposal-vs-report distinction, what is each one's time orientation? (proposal = future, "we will..."; report = past, "we found...")
6. What is the viva voce? (the oral examination/defence at which a dissertation candidate defends their work before examiners)