{"id":1344,"date":"2026-09-10T10:47:56","date_gmt":"2026-09-10T01:47:56","guid":{"rendered":"https:\/\/www.mindware-jp.com\/en\/?page_id=1344"},"modified":"2026-09-10T12:28:43","modified_gmt":"2026-09-10T03:28:43","slug":"what-is-a-concept-research","status":"publish","type":"page","link":"https:\/\/www.mindware-jp.com\/en\/what-is-a-concept-research\/","title":{"rendered":"What is a Concept Research?"},"content":{"rendered":"\n

The Myth of “Gathering Information Exhaustively and Analyzing It Using the Correct Method”<\/h2>\n\n\n\n

When exploring themes for R&D, new product\uff53, or new business, one common approach involves listing a multitude of keywords and then systematically and exhaustively investigating the details of each according to a set format. While this may appear rational at first glance, is it truly rational? To put it bluntly, it is a form of bureaucratic rationality.<\/p>\n\n\n\n

Herbert Simon discussed the concepts of complete versus bounded rationality and procedural versus substantive rationality. Although these were presented in separate works, if I may take the liberty of integrating them, the result is the following chart:<\/p>\n\n\n\n

\"\"<\/figure>\n\n\n\n

To engage effectively with the complexities of the real world, we ought to question whether the notion of “perfect rationality” is merely a fiction. Scrutinizing every piece of information in existence is simply beyond human capability. Yet, some may remain unshaken in the belief that by investigating more keywords and more items, they can\u2014if not achieve perfection\u2014at least draw closer to it. In many cases, however, what awaits them is nothing but exhaustion; for the world is vast<\/strong>.<\/p>\n\n\n\n

To draw an analogy with machine learning, complete rationality corresponds to a brute-force approach\u2014that is, a method of examining every possible case. This is feasible when the search space is small. However, if the search space is vast and only a portion of it is explored, one fails to reach the optimal solution and instead gets trapped in a local optimum. To address this, various shortcuts have been devised; these are known as heuristics.<\/p>\n\n\n\n

No matter how detailed the information you gather, if you limit your search to a narrow area, you are unlikely to find that “treasure.” What kind of treasure are you looking for? What are its characteristics? In short, that is what “conceptual research” entails. It is called “conceptual research” to distinguish it from “factual research.” The idea is to examine the concept\u2014its extension and intension (the set and its shared characteristics)\u2014before searching for the facts themselves.<\/p>\n\n\n\n

Searching for things you don’t yet know<\/h2>\n\n\n\n

As for the origins of this concept, I actually devised it based on my past experience planning and managing corporate seminars at a publishing and seminar company. The business of such a company involves gathering attendees and collecting fees. The seminar topics themselves could be anything\u2014information and communications, biotechnology, chemistry, energy, and so on. However, there were certain patterns common to seminars that attracted large audiences. In other words, rather than starting with a keyword and researching it, the approach was to start with a “winning pattern<\/strong>” and then find a topic that fit that model.<\/p>\n\n\n\n

There is simply not enough time to build up specialized knowledge in every field from scratch. Rather than waiting to know everything before taking action, the only way to uncover the unknown in unfamiliar territory is to explore using abstracted, generalized structures<\/strong>, relationships<\/strong>, models<\/strong>, and patterns<\/strong> as guides. It is about setting aside specific technical details and focusing solely on the essential principles\u2014asking, “What is actually happening here?” This mindset should also prove valuable when companies seek to break down the barriers of traditional specialized fields to explore new themes.<\/p>\n\n\n\n

For instance, when looking for a marriage partner, you cannot\u2014nor do you need to\u2014consider every single person of the opposite sex in the world; you simply need to know what kind of person you would like to marry. The same applies when a company seeks out themes for research and development. The goal is to discover a “gem,” not to exhaustively scrutinize every piece of information<\/strong>. In many research settings, however, this inversion of ends and means occurs, making matters unnecessarily complicated. The root cause is a form of organizational rationality known as “procedural rationality.” While this is unavoidable when working within an organization, one should keep in mind that it can easily drift far away from “substantive rationality.”<\/p>\n\n\n\n

In the past\u2014before the advent of LLMs\u2014attempting to explain the concept of “concept research” often led to overly philosophical discussions. This was because explaining the process to others, with the assumption that the work would be done manually, required breaking down the underlying thought processes. Now, however, we have LLMs. Since the essence of an LLM lies in language itself, it is capable of manipulating concepts; by leveraging LLMs and machine learning, we have been able to transform this into a practical technical tool.<\/p>\n\n\n\n

Let me give you an example of a simple experiment. Try asking an LLM questions such as, “Find applications for wedge engagement other than fastening,” or “In what situations might the principle of stably and uniformly dispersing fine particles over a long period be useful?” You should receive a wide range of answers. LLMs have been trained on technical information from all over the world. What is required is simply to formulate such questions based on your company’s own technological information.<\/p>\n\n\n\n

\"\"<\/figure>\n\n\n\n

Roadmap for the AI Industry and Enterprise Adoption<\/mark><\/strong><\/p>\n\n\n\n

<\/p>\n","protected":false},"excerpt":{"rendered":"

The Myth of “Gathering Information Exhaustively and Analyzing It Using the Correct Method” When exploring themes for R&D, new product\uff53, or new business, one common approach involves listing a multitude of keywords and then systematically and exhaustively investigating the details of each according to a set format. While this may appear rational at first glance, […]<\/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-1344","page","type-page","status-publish","hentry"],"_links":{"self":[{"href":"https:\/\/www.mindware-jp.com\/en\/wp-json\/wp\/v2\/pages\/1344","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.mindware-jp.com\/en\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/www.mindware-jp.com\/en\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/www.mindware-jp.com\/en\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.mindware-jp.com\/en\/wp-json\/wp\/v2\/comments?post=1344"}],"version-history":[{"count":5,"href":"https:\/\/www.mindware-jp.com\/en\/wp-json\/wp\/v2\/pages\/1344\/revisions"}],"predecessor-version":[{"id":1362,"href":"https:\/\/www.mindware-jp.com\/en\/wp-json\/wp\/v2\/pages\/1344\/revisions\/1362"}],"wp:attachment":[{"href":"https:\/\/www.mindware-jp.com\/en\/wp-json\/wp\/v2\/media?parent=1344"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}