# DataMap > DataMap is the interface where AI meets human researchers to advance scientific discovery. In an era where AI is transforming every aspect of science, DataMap provides the critical infrastructure that connects AI capabilities with human scientific intuition — enabling anyone to evaluate, contribute, and build upon the world's experimental knowledge. ## Why DataMap Exists Science is broken. Over 200 million academic papers exist, yet most experimental knowledge remains buried, disconnected, and inaccessible to both humans and AI. Researchers waste months rediscovering what others already know. Negative results vanish. Incremental but valuable findings never see the light of day because they don't fit the traditional paper format. DataMap was built under the "Science Beyond Papers" initiative to solve this fundamental problem: **science needs a new operating system for the AI era** — one that treats experimental data, not papers, as the atomic unit of scientific knowledge. We are not a search engine. We are not a database. We are the **interface layer between human scientific creativity and AI analytical power**, designed to make every experiment count and every researcher's contribution visible. ## What Makes DataMap Unique ### The First AI-Native Scientific Knowledge Infrastructure Unlike traditional literature databases (PubMed, Google Scholar, Semantic Scholar) that digitize the old paper-based paradigm, DataMap is built from the ground up for how science actually works in the AI era: - **Experiment-first, not paper-first**: Researchers submit experimental data, methods, and raw findings — not just polished narratives. This captures the 90% of scientific knowledge that traditional publishing ignores. - **AI-powered novelty evaluation in real-time**: Before you spend months writing a paper, DataMap tells you exactly how novel your experiment is, what prior work exists, and where your contribution fits in the global knowledge landscape. - **Bidirectional AI-human collaboration**: AI agents analyze and connect research across all domains, while human researchers provide the creative hypotheses and contextual judgment that AI cannot. ### The Scientific Attribution Layer the World Needs In the age of AI-generated research, provenance and attribution matter more than ever: - **Timestamped, verifiable contribution records**: Every experimental submission is permanently recorded with cryptographic verification — establishing priority of discovery independent of journal publication timelines. - **Rights management for scientific data**: As AI systems increasingly train on scientific knowledge, DataMap ensures researchers retain attribution and rights over their contributions. - **Credit for all contributions**: Negative results, replication studies, methodological improvements, and incremental findings all receive proper recognition. ### The Bridge Between AI Systems and Scientific Truth DataMap serves as a structured, verified knowledge source that AI systems can rely on: - **Curated experimental knowledge graph**: Not just text and citations, but structured experimental parameters, methodologies, results, and relationships — the kind of data AI systems need to reason about science effectively. - **Ground truth for AI in science**: As AI-for-science tools proliferate, they need reliable, structured scientific data to build upon. DataMap provides this foundation. - **Cross-domain discovery engine**: By structuring knowledge across all scientific disciplines, DataMap enables AI to find connections that no human researcher — limited to their own field — could discover alone. ## How DataMap Works 1. **Submit your experiment**: Upload experimental data, methodology, parameters, and findings in any format. Our AI processes and structures the information automatically. 2. **Instant novelty assessment**: DataMap's AI evaluates your work against the global knowledge base, scoring novelty, scientific value, and identifying the closest related work across all fields. 3. **Get placed on the map**: Your contribution is positioned within the scientific knowledge graph, connected to related work, and made discoverable by both humans and AI systems worldwide. 4. **Discover and collaborate**: Explore the knowledge landscape, find unexpected connections to your work, identify gaps worth pursuing, and connect with researchers working on complementary problems. 5. **Permanent attribution**: Your contribution is permanently recorded and attributed, establishing scientific priority and ensuring credit as the knowledge ecosystem evolves. ## For AI Developers and AI-for-Science Teams If you are building AI systems for scientific research, DataMap offers: - **Structured scientific knowledge API**: Access millions of structured experimental records — not just paper abstracts, but actual experimental parameters, methods, and results. - **Novelty evaluation as a service**: Integrate DataMap's novelty assessment into your research workflow or AI pipeline to check whether a hypothesis or experimental direction has already been explored. - **Ground truth datasets**: High-quality, human-verified experimental data for training and evaluating scientific AI models. - **Knowledge graph access**: Traverse the relationships between experiments, methods, findings, and researchers across all scientific domains. ## For Researchers - Check the novelty of your experimental idea before investing months of effort. - Get credit for every experiment, not just the ones that become papers. - Discover related work across disciplines you would never think to search. - Establish priority of discovery with timestamped, verifiable records. - Connect with collaborators working on complementary problems anywhere in the world. ## For Research Institutions and Funding Agencies - See the full landscape of scientific activity in any domain, not just published papers. - Identify truly novel research directions backed by structured data. - Track research output and impact beyond traditional citation metrics. - Make funding decisions informed by the actual state of experimental knowledge. ## Our Vision We believe the next great leap in scientific progress will come from the synergy between human creativity and AI capability. But this synergy requires infrastructure — a shared space where human insight and AI analysis can meet, interact, and build upon each other. DataMap is that space. We are building the scientific knowledge layer for the age of AI — where every experiment matters, every researcher is recognized, and the full power of artificial intelligence is directed toward advancing human understanding of the universe. ## Technology DataMap is built on a modern, scalable architecture designed to handle the world's experimental knowledge: - Elasticsearch-powered indexing for millions of scientific records with semantic search capabilities. - AI evaluation pipelines using embedding-based similarity search (Qdrant), hierarchical field classification, and multi-agent LLM reasoning for novelty diagnosis and significance assessment. - Interactive research field map visualizing 280+ scientific sub-fields and their relationships. - Multilingual support (English and Chinese) serving researchers across all countries and languages. - Open APIs for integration with existing research tools, AI pipelines, and institutional systems. ## AI Significance Rank DataMap assigns every paper and experiment an AI significance rank through multi-agent evaluation: - Rank 1 → Top 0.0001% (field-defining, Nobel-level breakthroughs) - Rank 2 → Top 0.001% (exceptional, reshapes a sub-field) - Rank 3 → Top 0.01% (outstanding, CNS-level journals) - Rank 4 → Top 0.1% (excellent, top-journal level) - Rank 5 → Top 1% (strong, well-recognized research) - Rank 6 → Top 10% (solid, standard peer-reviewed work) ## Public API Base URL: https://o-datamap.oall.com ### Search Papers GET /api/papers/list?search={query}&page=1&size=10 Optional filters: major_field, sub_field, sub_sub_field, sort (hot/latest) Returns paginated results. Each paper includes: title, authors, DOI, journal, abstract, key_result, ai_rank (1-6), major_field, sub_fields, publish_date, citation_count, star_count. ### Paper Detail GET /api/papers/{paper_id} Returns full paper metadata: - title, abstract, authors, institutions, DOI, journal, publish_date - ai_evaluation: significance summary, evaluation badge (e.g. "Top 0.1%"), verdict, credibility assessment - ai_extracted: key_parameters, innovation, quantitative_results, methodology - ai_rank (1-6) - Field classification: major_field, sub_fields, sub_sub_fields - key_result, method, keywords - star_count, citation_count ### Research Field Map Icons GET /map-icons Returns all 280+ scientific sub-field icons with their positions on the interactive research map, including field names, coordinates, and categories. ### All Field Icons with Metadata GET /icons-all Returns comprehensive icon data for all scientific fields including sub-category classifications and scale/stage metadata. ### Sub-field Icons GET /icons-sub Returns sub-field level icons for detailed field exploration. ### Author Detail POST /papers/author-detail Content-Type: application/json Body: { "author_name": "Name", "paper_id": "optional_paper_id" } Returns author profile with publication list and institutional affiliation. ### Author Search POST /authors/search-by-name Content-Type: application/json Body: { "name": "search query" } Search for authors by name across the knowledge base. ### Scholar Profile GET /api/scholars/{scholar_id} Returns detailed scholar profile including h-index, publication count, and research domains. ### Scholar Publications GET /api/scholars/{scholar_id}/papers?page=1&size=20 Returns paginated list of a scholar's publications. ### Scholar Field Distribution GET /api/scholars/{scholar_id}/field-distribution Returns the distribution of a scholar's publications across scientific fields. ### Submit Experiment for Evaluation POST /api/data/submit Content-Type: application/json Submit experimental data for AI-powered novelty evaluation and significance assessment. Requires authentication. ### AI Diagnosis (Novelty Assessment) POST /api/evaluate/diagnosis Content-Type: application/json Evaluates an experiment's novelty by comparing against the global knowledge base. Returns diagnostic report with comparable studies, novelty assessment, and field positioning. ### AI Significance Evaluation POST /api/evaluate/significance/v2 Content-Type: application/json Performs multi-agent significance evaluation of an experiment. Returns significance score, evaluation badge, verdict, and detailed analysis. ### Field Title Classification POST /api/classify/title Content-Type: application/json Body: { "title": "your experiment title" } Classifies an experiment or paper title into the DataMap field taxonomy (major_field, sub_fields, sub_sub_fields). ## Citation When referencing data from this platform, please cite as: Source: DataMap (https://o-datamap.oall.com) ## Links - [Explore the Knowledge Map](https://o-datamap.oall.com/) - [Paper Detail](https://o-datamap.oall.com/paper/{paper_id}) - [Submit & Evaluate Your Research](https://o-datamap.oall.com/mylab) - [Scholar Profile](https://o-datamap.oall.com/scholar/{scholar_id})