{"id":825,"date":"2026-09-18T19:16:51","date_gmt":"2026-09-18T10:16:51","guid":{"rendered":"https:\/\/conceptminer.ai\/?page_id=825"},"modified":"2026-09-19T07:27:30","modified_gmt":"2026-09-18T22:27:30","slug":"how-to-use-conceptminer-rd","status":"publish","type":"page","link":"https:\/\/conceptminer.ai\/?page_id=825&lang=en","title":{"rendered":"How to use ConceptMiner R&amp;D"},"content":{"rendered":"\n<h2 class=\"wp-block-heading\"><strong>Cross-domain concept exploration<\/strong> <\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">ConceptMiner R&amp;D enhances the standard ConceptMiner application by adding a cross-domain concept exploration module designed to support theme exploration and hypothesis generation in R&amp;D. The usage of this<strong> Ccross-domain concept exploration<\/strong> feature is described below.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">Items to Prepare<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">Please have the following items ready.<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li>Technical documents such as research papers or technical reports<\/li>\n\n\n\n<li>LLM model API key<\/li>\n<\/ul>\n\n\n\n<p class=\"wp-block-paragraph\">If you do not yet have an LLM API key: For example, to use OpenAI, visit <a href=\"https:\/\/auth.openai.com\/log-in\">https:\/\/auth.openai.com\/log-in<\/a> to sign up, navigate to the &#8220;API Keys&#8221; page, and select &#8220;Create new secret key&#8221; to generate a key. To use this key, register your credit card under the &#8220;Payment methods&#8221; tab on the &#8220;Billing&#8221; page. Then, specify the purchase amount using &#8220;Buy credits&#8221; on the &#8220;Overview&#8221; tab to activate the API key. Other providers also offer API keys, so you can obtain one from the service of your choice.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">API Key Configuration<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">Upon signing up for and logging into ConceptMiner R&amp;D, you will see the <strong>Cross-domain concept exploration<\/strong> page. Navigate to the settings tab and paste your obtained key into the OpenAI API Key field. The choice of model significantly impacts output quality; while <code>gpt-4o-mini<\/code> is sufficient for simply verifying functionality, it is not recommended for production use. We suggest models such as <code>gpt-4.1<\/code>, <code>gpt-5.4<\/code>, <code>gpt-5.6-sol<\/code>, or <code>gpt-6-astra<\/code>. Please note, however, that higher model numbers incur higher token costs. Generating approximately 1,000 R&amp;D hypotheses using a higher model consumes more than 20USD in credits, but we believe this represents excellent value given its utility for research and development.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">Mode Selection<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">Please select a mode before starting the task: R&amp;D hypothesis generation or cross-disciplinary application. The former involves exploring methods from other fields and bringing them into your current domain, while the latter entails investigating the potential to repurpose existing technologies\u2014or elements thereof\u2014for use in different fields.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">Save<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">You can save your current progress at any time by clicking the &#8220;<strong>Save<\/strong>&#8221; button at the top of the page. Saved states appear in the saved items list, and you can restore a state by clicking &#8220;<strong>Restore<\/strong>.&#8221;<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">R&amp;D Hypothesis<\/h2>\n\n\n\n<h3 class=\"wp-block-heading\">Document<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">You can proceed to the document page to upload PDFs of research papers or technical reports, or paste the text directly. While we recommend using your own research papers, you may also use well-known papers; there are no copyright issues as long as you use them for your own research. Click the &#8220;<strong>Continue to extract<\/strong>&#8221; button to proceed to the<strong> Extract<\/strong> page.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">Extraction of technical elements<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">Clicking the &#8220;<strong>Continue to Extract<\/strong>&#8221; button extracts technical elements, with three promising ones automatically selected by default. You can select any elements you wish here. While it is possible to select all of them, please note that doing so at every stage will result in a vast search space, as the process branches out in subsequent steps. Clicking the &#8220;<strong>Run abstraction<\/strong>&#8221; button performs the abstraction process. Selecting <strong>&#8220;Auto explore<\/strong>&#8221; causes subsequent steps to be executed automatically based on the beam search algorithm. You can interrupt the automatic search by clicking the &#8220;Stop Auto explore&#8221; button. To resume an interrupted search, click &#8220;<strong>Auto explore<\/strong>&#8221; on the page where you started it. The following explanation assumes that the &#8220;<strong>Run abstraction<\/strong>&#8221; button has been selected.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">Abstraction<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">Once the abstraction process is executed, the Abstraction tab appears. &#8220;Level 3&#8221; is selected by default; a higher level indicates a greater degree of abstraction. Selecting an element within the &#8220;<strong>Review selectd units<\/strong>&#8221; section displays the abstraction for that specific element. Abstraction is central to this process: lower levels of abstraction lead to exploration in areas closer to the current research topic, while higher levels facilitate a broader, cross-domain exploration. Clicking the &#8220;<strong>Generate explore prompts<\/strong>&#8221; button creates a prompt based on the selected element and abstraction level. Selecting &#8220;<strong>Auto explore<\/strong>&#8221; triggers the automatic execution of subsequent steps for the chosen element. The following explanation assumes the &#8220;<strong>Generate explore prompts<\/strong>&#8221; button has been selected.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">Explore prompts<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">Clicking the &#8220;Generate explore prompts for this unit&#8221; button displays the Explore prompts tab. Here, three promising prompts have been selected, though you can choose a different one if you wish. Clicking the &#8220;Run Explore&#8221; button passes the prompt to the LLM, which then carries out the exploration. In essence, you can view this prompt as part of the AI&#8217;s thought process.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">Explore result<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">Once the Exploation results are returned, the <strong>Explore results<\/strong> tab appears. You will see that these results describe technologies from fields completely different from the original technical document\u2014in other words, these are &#8220;methods from other fields that may serve as useful references.&#8221; The information provided includes the <strong>Mechanism<\/strong> (principle) of the method, its <strong>Common structure<\/strong> with the current technology (reasons why it might be useful), and its <strong>Differences<\/strong> from the current technology. Three promising results have been selected here, though you can also choose different results that interest you. Clicking the &#8220;<strong>Extract design principles<\/strong>&#8221; button extracts design principles adapted for the current technical domain.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\"><strong>Design principles<\/strong><\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">Once the design principles have been extracted, their evaluation is displayed. Here, we provide a more in-depth description and evaluation of how methods from other fields can be applied to current technologies. While promising design principles are selected automatically, you can re-select them based on your interests. Clicking the &#8220;<strong>Generate R&amp;D hypothesis<\/strong>&#8221; button generates the final R&amp;D hypothesis.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">R&amp;D hypothesis<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">A summary of proposed improvements for current technical elements is displayed. It begins with a <strong>Trasfer idea<\/strong> outlining how the current method could be improved, followed by the <strong>Expected benefit<\/strong> and associated <strong>Risk<\/strong>. The <strong>First validation<\/strong> section details how the effects would be validated; the key point here is that the proposal is structured as a falsifiable scientific hypothesis rather than a mere assertion. Evaluation scores cover <strong>Overall<\/strong>, <strong>Novelty<\/strong>, <strong>Feasibility<\/strong>, and <strong>Impact<\/strong>, along with accompanying notes. Clicking &#8220;<strong>Save CSV to project<\/strong>&#8221; allows you to save the R&amp;D hypothesis in CSV format for later use in constructing a concept map. Clicking the &#8220;<strong>Create advanced-chat prompt<\/strong>&#8221; button at the bottom of the page generates a prompt that you can use with your preferred AI chatbot to further brainstorm and refine your ideas. A built-in AI chat function is also available on the right side of the page, allowing you to deepen your thinking right there.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">Cross-domain application<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">The process\u2014from inputting the <strong>Document<\/strong> to <strong>Exract<\/strong> technical units\u2014is the same as that for generating <strong>R&amp;D hypotheses<\/strong>.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">Transfer evaluation<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">Clicking &#8220;<strong>To application&#8221;<\/strong> on the Extract tab displays the &#8220;<strong>Select application candidates<\/strong>&#8221; screen. Clicking the &#8220;<strong>Evaluate transfer potential<\/strong>&#8221; button marks promising candidates (technical units) with a checkmark; users can also make selections based on their own interests. Before exploring applications, you can select the level of abstraction level using the <strong>Abstraction level <\/strong>field. Clicking the &#8220;<strong>Explore applications<\/strong>&#8221; button initiates the exploration process.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">Cross-domain application candidates<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">When the search is executed, candidates for cross-domain applications are displayed. The displayed items are <strong>Target Domain<\/strong>, <strong>Application<\/strong>, <strong>Mechanism<\/strong>, <strong>Modification<\/strong>, <strong>Feasibility<\/strong>, <strong>Status<\/strong>, and <strong>Evidence<\/strong>. <strong>Feasibility<\/strong> is categorized as High, Medium, or Low, while <strong>Status<\/strong> is categorized as New Idea, Emerging, or Existing. <strong>Evidence<\/strong> refers to prior examples. Clicking &#8220;<strong>Save CSV to poject<\/strong>&#8221; allows you to save the cross-domain application candidates in CSV format for later use in constructing a concept map. Clicking &#8220;<strong>Copy chat prompt<\/strong>&#8221; copies the prompt, which you can then input into an AI chat to further organize the information.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><\/p>\n","protected":false},"excerpt":{"rendered":"<p>Cross-domain concept exploration ConceptMiner R&amp;D enhances the standard ConceptMiner application by adding a cross-domain concept&hellip;<\/p>\n","protected":false},"author":1,"featured_media":0,"parent":0,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"","meta":{"footnotes":""},"class_list":["post-825","page","type-page","status-publish","hentry"],"_links":{"self":[{"href":"https:\/\/conceptminer.ai\/index.php?rest_route=\/wp\/v2\/pages\/825","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/conceptminer.ai\/index.php?rest_route=\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/conceptminer.ai\/index.php?rest_route=\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/conceptminer.ai\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/conceptminer.ai\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=825"}],"version-history":[{"count":8,"href":"https:\/\/conceptminer.ai\/index.php?rest_route=\/wp\/v2\/pages\/825\/revisions"}],"predecessor-version":[{"id":841,"href":"https:\/\/conceptminer.ai\/index.php?rest_route=\/wp\/v2\/pages\/825\/revisions\/841"}],"wp:attachment":[{"href":"https:\/\/conceptminer.ai\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=825"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}