Material Phase 2: Proposal Writing & Research Design

1. Topic Generation & The FINER Criteria

A research proposal is a formal document advocating for the execution of a study. Before writing, a candidate CS research topic must be evaluated using the FINER Criteria.

F - Feasible

Is the study achievable within the available timeframe (e.g., final-year semester), budget, computational hardware (GPUs, servers), and data access constraints?

I - Interesting

Does the topic intrigue both the researcher and the broader computing community, addressing an active industry or academic priority?

N - Novel

Does it provide new insights, extend an existing algorithm to a new context, build a novel system artifact, or re-evaluate existing findings with modern benchmarks?

E - Ethical

Does the research respect user privacy, data security, and institutional ethics guidelines (no unauthorized data scraping or security exploits without authorization)?

R - Relevant

Does it contribute to advancing CS knowledge, solving real-world infrastructure problems, or improving software practices?

2. Anatomy of a NUST CS Research Proposal

In SCS2110 examinations and final-year project defenses, a research proposal must include the following compulsory sections:

Descriptive Title: Concise, precise, omitting fluff words like "A Study of...". Must state variables/technology and target context.
Background & Motivation: Introduces the technological context, historical trends, and sets up the problem.
Problem Statement: Explicitly cites empirical literature to prove the existence of an unresolved gap.
Aim and Objectives: Overarching destination (Aim) + 3-4 SMART action-verb steps (Objectives).
Justification & Significance: Explains why this research matters to academia, industry, or policy makers.
Scope & Delimitations: Defines what is strictly included and excluded from the study boundary.
Preliminary Literature Review: Synthesizes existing state-of-the-art solutions and highlights the knowledge gap.
Proposed Methodology: Choice of paradigm, artifact architecture, data collection techniques, and evaluation metrics.
Work Plan & Timeline (Gantt Chart): Visual schedule breaking tasks into milestones.
Resource & Budget Requirements: Hardware, cloud API credits, dataset license costs.
Ethical Considerations: Protocol for handling human subjects or sensitive source code data.
References (Harvard Style): Complete list of cited literature sources.

🎯 Exam Trap Alert: Scope vs. Delimitation

Scope defines the boundaries of the study set by the research objectives (e.g., "This study evaluates network latency only on IPv6 campus wireless networks").
Delimitations are parameters intentionally excluded by the researcher to keep the project feasible (e.g., "5G cellular traffic is intentionally excluded due to lack of local hardware access").

3. Research Design Frameworks in CS

A research design is the master blueprint specifying how data will be collected, measured, and analyzed to answer the research questions.

Design Framework	Primary Purpose	Typical CS Application
Exploratory Design	Used when little is known about a new domain or emerging technology. Focuses on discovering insights.	Investigating developer adoption challenges for WebAssembly or quantum computing frameworks.
Descriptive Design	Accurately profiles the characteristics of a system, population, or phenomenon without manipulating variables.	Measuring current cybersecurity vulnerability patch distribution rates across municipal networks.
Explanatory Design	Seeks to explain cause-and-effect relationships between variables.	Determining why database query compilation time increases exponentially with nested join depth.
Experimental Design	Manipulates an independent variable under controlled conditions to observe changes in a dependent variable.	Benchmarking execution speed of a novel lock-free concurrent queue against standard Pthreads mutex queues under load.

4. Feasibility, Risk Management & Timeline (Gantt Chart)

Evaluators scrutinize proposals for practical viability. A proposal without risk mitigation or realistic timelines will be rejected.

Technical Feasibility

Verifies whether required algorithms, APIs, SDKs, or hardware models are accessible and computationally tractable.

Operational Feasibility

Ensures target users or sample groups (e.g., network admins, beta testers) are reachable and willing to participate.

Risk Mitigation Matrix

Identifies potential project pitfalls (e.g., loss of cloud access, API deprecation) and specifies fallback contingency plans.

💡 Distinction Rule for Gantt Charts:
Always ensure research proposal Gantt charts display sequential dependencies (e.g., Literature Review must complete before Final System Architecture Design, and Data Collection must precede Statistical Analysis).

Material Phase 2: Research Proposal Writing & Design Frameworks