Thesis Atlas is an AI-powered system that helps you discover research title candidates that are relevant, focused, and realistic, ranked using multi-dimensional scoring (Novelty, Impact, Feasibility, and Duplication Risk) with clear, easy-to-understand explanations.

Thesis Atlas is designed for D3/D4, Bachelor’s, Master’s, and PhD students who:

• are still unsure about choosing a topic/title,
• want to reduce the risk of duplicated titles,
• need titles that fit their time, data access, and method constraints,
• want a solid rationale for “why this title is worth pursuing.”

A simple guided flow:

• You provide context (interests, faculty/program, keywords, available time, data access, etc.).
• The system generates multiple title candidates and ranks them with a 4D score.
• You can discuss each title with AI to sharpen the scope, method, and contribution.
• Save a shortlist and use it as a structured basis for supervisor review.

• Novelty: how new/unique the idea is compared to existing references.
• Impact: the potential academic/practical contribution (significance and relevance).
• Feasibility: how realistic it is to complete (time, data access, complexity).
• Duplication Risk: the likelihood the title is too similar to existing work.

The score is a decision-support tool, not a replacement for your supervisor. Thesis Atlas helps you compare candidates consistently and transparently, but the final decision should follow your supervisor’s guidance, program policies, and source verification.

No system can guarantee 100%. However, Thesis Atlas helps reduce the risk through similarity checks and duplication signals, so you can avoid titles that are too close to existing work early on.

Recommendations are shown in Indonesian by default. You can discuss them with AI to generate an English version (or variations).

Recommended inputs:

• 3–5 specific keywords
• Clear context (faculty/program, field, case setting)
• Time constraints (e.g., 1–2 semesters)
• Data access preference (internal, public, survey, etc.)

Yes. Just write your interest or a short description (e.g., “using AI agents in office administration”), and let the system suggest directions and title candidates.

Yes. Thesis Atlas can:

• refine/sharpen your title,
• propose stronger alternative variations,
• help narrow the scope and suggest methods,
• provide a rationale you can bring to your supervisor.

Yes. You can choose the research type (e.g., experimental, descriptive, ethnographic, qualitative, quantitative, etc.), and the system adapts recommendations and scoring based on your context and constraints.

• Data access: how easily you can obtain the data (e.g., internal institutional data, online surveys, secondary datasets, etc.).
• Data source: where the data comes from (open/public data, institutional internal data, proprietary/licensed datasets, API/platform data, primary collection, mixed).

Yes. You can choose options like open/public data, primary data collection (survey/interview), or platform/API data. The system will suggest titles that are realistic given your access constraints.

Not automatically. If you choose “Platform/API data,” the system provides general guidance that respects platform Terms of Service (ToS) and considers constraints such as rate limits and data formats.

Yes. The system can suggest suitable methods (surveys, case studies, experiments, document analysis, mixed methods, etc.) based on your context, data access, and available time.

Google/repositories mostly return raw results. Thesis Atlas helps you:

• incorporate your personal constraints (time, data access, program),
• compare candidates using multi-dimensional scores with clear explanations,
• reduce duplication risk through similarity signals,
• move faster from an “idea” to a research-ready title.

We store limited input and usage activity to improve recommendation quality, enhance system performance, and develop Thesis Atlas models/features. This data is used solely for service improvement purposes, and we do not sell it to third parties.

Your conversations are protected with encryption in transit and at rest, and access is restricted under our system policies. We apply access controls and security practices to help keep your data confidential.

Yes. You can use cross-domain keywords, and the system will balance relevance with your faculty/program selection (if provided) so the recommendations remain aligned.

Add more concrete details such as:

• setting (location/institution/target group),
• object of study (application, policy, system),
• constraints (time period, dataset, key variables),
• goal (evaluation, prediction, recommendation, comparison).

Each recommendation comes with a short, human-readable explanation: novelty signals, relevance cues, duplication-risk indicators, and feasibility factors, so you understand why the title is suggested and what to do next.

Yes. You can build a shortlist, save promising candidates, and return anytime to continue discussion or refine them further.

Exporting (e.g., PDF/CSV) is not available yet. Coming soon.

Yes. Thesis Atlas is free to use with two modes: Guest and Registered.

• Guest: up to 3 title generations and AI Chat up to 5 messages.
• Registered: full access to features, session history, and exploration without those limits. For full access, please register.

No. Thesis Atlas accelerates exploration and helps you prepare stronger options for discussion, but final decisions must follow your supervisor’s guidance and your program’s policies.

Yes. You can set your preference (e.g., prioritize feasibility), and the system will balance recommendations to stay realistic while still aiming for strong contribution.

Refine your faculty/program selection, add clearer keywords, and use AI discussion to “lock” the scope. The system can also constrain recommendations to the selected domain.

After choosing a candidate title, you can typically save it, claim it, and (if needed) submit it for admin verification. To prevent title collisions across users, Thesis Atlas uses 8 global title statuses:

• AVAILABLE: No meaningful activity from other users yet. Free to explore.
• VIEWING: Another user is currently viewing it (based on recent heartbeat). Not locked.
• DISCUSSING_LOW: discussion occurred in the last 24 hours, but at low intensity.
• DISCUSSING_HIGH: High AI discussion in the last 24 hours (≥ threshold; currently 5 chats).
• SAVED: Added to at least one user’s shortlist. Not exclusive.
• CLAIMED: Exclusively claimed by a user for a limited period. Others cannot claim, and saving is disabled while active.
• SUBMITTED: Submitted for admin verification. Cannot be claimed by other users.
• REGISTERED: Approved by admin and officially registered. Final status.

Practical tip: If a title is still non-final (AVAILABLE/VIEWING/DISCUSSING/SAVED), you can usually refine it (scope, variables, context, method) to make it more unique and stronger.

Yes. The status system helps prevent collisions across users through a claim → submit → admin verification workflow (as designed in the system).

Please reach out via: Contact Me