Software patents have their own characteristics compared to patents in other technical fields. The biggest feature is that technical solutions often cannot be completely presented from the outside. For example, when it comes to algorithms, the technical solutions often exist inside the computer. When it comes to process processing, technical solutions often require the participation of multiple parties. In other words, software patent technical solutions usually cannot be fully perceived by users based on the product. Because of this, to do a good job in mining software patents, the author believes that it can be done from the following two aspects.
The guidance of R&D personnel is particularly important. Since the mining of software patents is usually based on a core technical solution or a product (platform, application, etc.) R&D project, the mining personnel first need to guide the R&D personnel to describe the core technical solution or the R&D project itself as a whole. Specifically include:
1) Software application scenarios and target users
Since software can be applied in all walks of life, the software itself can be subdivided into fields, so it is first necessary to guide research and development The personnel describe the application scenarios and target users of the software. For example, the software is used in the payment field, logistics field, smart home appliance field, etc., and whether the target users of the software are mass users or specific users (such as office lobby staff, bank staff or salesmen, etc.).
2) The main functions and highlights of the product
Guide the R&D personnel to describe the main functions and highlights of the product in order to clarify "what this software can do" and "what this software has done" "Things are different from other software." The product highlights can be used to locate important innovation points when subsequently digging for specific innovation points. Generally speaking, R&D personnel will describe the main functions and highlights of the product based on their understanding of their peers. Therefore, after discovering innovation points, it is also necessary to search existing technologies and sort out important innovation points again based on existing technologies.
2. Technical decomposition
After having a certain understanding of the core technical solution and the project itself as a whole, in order to find out the innovation points, a "big" software needs to be disassembled. It is divided into "small" parts that can be used to explore technological innovation points. Specifically include:
1) Functional disassembly
This disassembly method is the most common for software patent mining, because most software can be disassembled based on functions Division, and the technical solutions of software patents are generally described from a functional perspective. When dismantling functionally, it can be disassembled step by step until the functional modules are minimized. Then start from different functional modules and guide the R&D personnel to describe the innovation points. For example, does this function exist in similar products? If it does not exist, it is a new function. If it exists, further guide whether there is any difference in the implementation of the function.
2) Dismantle the technical architecture
Software needs to run in a computer system. For computers, technicians usually understand the logical structure and structure of the computer through the computer architecture. functional characteristics. For example, for a computer system, it can be divided into basic application layer, basic platform layer and data layer. This vertical structure is often familiar to technicians. Therefore, when digging for innovation points, we can also split it from the technical architecture, and guide R&D personnel to describe the innovation points from each level, so as to accurately find the innovation points of software patents.
Disassembled from the business process
Many software applications are used to implement various business processes. For business processes, they are usually divided into multiple stages, so making software When mining patents, the overall business process can be divided into multiple stages, and R&D personnel can be guided from different stages to describe the innovation points, thereby finding the innovation points of software patents.
The above only shows a few common methods in software patent mining, but of course it is not limited to these methods. It should be noted that when mining software patents, it is not enough to just use one of them. In fact, in the process of patent mining, various methods should be integrated and multiple methods should be used. The purpose is to achieve comprehensive mining. For example, when dismantling the technical architecture, for the application layer, it is necessary to further guide R&D personnel to divide it from a functional perspective and find innovation points for each function. Moreover, after sorting out the innovation points in each link, you should also pay attention to the correlation between different innovation points, for example, whether different innovation points are based on solving the same technical problem.
Therefore, after sorting out the innovation points, you should also consider whether you need to combine multiple innovation points into one application. In addition, during the mining process, more divergent thinking should be conducted on the unearthed technological innovation points to find some related technical factors, so as to find other possible technological innovation points. Based on the excavated innovation points, a patent mining list will eventually be formed