Automatic competitive product analysis method and terminal
By constructing a training dataset and training a competitor feature analysis model, competitor information is analyzed automatically, solving the problems of low efficiency and insufficient accuracy in traditional competitor analysis, and achieving efficient and accurate acquisition of competitor information and product improvement.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- FUJIAN TQ DIGITAL
- Filing Date
- 2024-11-04
- Publication Date
- 2026-05-08
AI Technical Summary
Traditional manual competitive analysis methods are time-consuming, resource-intensive, and unable to provide a quick and comprehensive understanding of competitor information. Furthermore, their analysis efficiency and accuracy are insufficient.
By collecting competitor information and constructing a training dataset for feature identification, a competitor feature analysis model is trained. The model is then used to automatically analyze the second target product to obtain the competitor with the highest matching degree and its feature identification, and an improvement plan is generated.
It improves the efficiency and accuracy of competitive analysis, reduces manpower and time costs, and enables the rapid acquisition of representative competitor information and the generation of targeted improvement plans, thereby enhancing product quality and competitiveness.
Smart Images
Figure CN121998673A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of computer applications, and in particular to an automatic analysis method and terminal for competitor products. Background Technology
[0002] In the market, in-depth competitor analysis is a key step for companies to formulate effective market strategies and enhance product competitiveness. Traditionally, this task is performed manually by collecting vast amounts of competitor data. However, this method has limitations: it is time-consuming, requiring significant human and material resources to collect and organize relevant information, cannot complete competitor data analysis quickly, and necessitates considerable time spent experiencing competitors to understand their features and design highlights. To overcome these limitations, AI-based competitor analysis and deconstruction methods have emerged. This method can automatically deconstruct the functional modules of competitors, use machine learning techniques to compare and analyze the functions and performance of these modules, and simultaneously analyze market feedback data to extract user reviews and suggestions. The aim is to accurately deconstruct competitors, intelligently compare functions, comprehensively evaluate performance, and efficiently analyze market feedback, thereby providing companies with valuable product improvement suggestions and market strategies. Summary of the Invention
[0003] The technical problem to be solved by the present invention is to provide an automatic analysis method and terminal for competitors, which can improve the comprehensiveness and accuracy of data processing, thereby improving the efficiency and accuracy of competitor analysis.
[0004] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows: An automated competitor analysis method, comprising the following steps: S1. Collect information on the first competitor of the first target product, and construct a training dataset after feature labeling the first competitor information; S2. Train a competitor feature analysis model based on the first target product and its corresponding training dataset; S3. Input the data of the second target product into the competitor feature analysis model to obtain the competitor with the highest matching degree and its feature identifier, and generate an improvement plan for the second target product based on the competitor and its feature identifier.
[0005] To solve the above-mentioned technical problems, another technical solution adopted by the present invention is as follows: An automated competitor analysis terminal includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the various steps of the aforementioned automated competitor analysis method.
[0006] The beneficial effects of this invention are as follows: This invention provides an automatic competitor analysis method and terminal. By collecting first competitor information for a first target product, and then identifying the features of the first competitor information to construct a training dataset, a data foundation is laid for subsequent training of a competitor feature analysis model, ensuring the reliability of the model's training. The competitor feature analysis model is trained based on the first target product and its corresponding training dataset. This model then automatically collects and analyzes competitor data based on the data of a second target product, obtaining the competitor with the highest matching degree to the second target product and its feature identifiers. This improves the comprehensiveness and accuracy of data processing, thereby increasing the efficiency and accuracy of competitor analysis. Finally, an improvement scheme for the second target product is generated based on the competitor and its feature identifiers, enhancing the design quality and competitiveness of the second target product. Attached Figure Description
[0007] Figure 1 This is a flowchart of an automatic competitor analysis method according to an embodiment of the present invention; Figure 2 This is a schematic diagram of an automatic analysis terminal for a competitor product according to an embodiment of the present invention; Figure 3 This is a flowchart illustrating the training process of the competitor feature analysis model according to an embodiment of the present invention. Label Explanation: 1. An automated analysis terminal for a competing product; 2. Memory; 3. Processor. Detailed Implementation
[0008] To explain in detail the technical content, objectives, and effects of the present invention, the following description is provided in conjunction with the embodiments and accompanying drawings.
[0009] Please refer to Figure 1 This invention provides an automatic competitor analysis method, comprising the following steps: S1. Collect information on the first competitor of the first target product, and construct a training dataset after feature labeling the first competitor information; S2. Train a competitor feature analysis model based on the first target product and its corresponding training dataset; S3. Input the data of the second target product into the competitor feature analysis model to obtain the competitor with the highest matching degree and its feature identifier, and generate an improvement plan for the second target product based on the competitor and its feature identifier.
[0010] As can be seen from the above description, the beneficial effects of the present invention are as follows: by collecting information on the first competitor of the first target product, and constructing a training dataset after feature identification of the first competitor information, a data foundation is laid for subsequent training of the competitor feature analysis model, ensuring the reliability of the competitor feature analysis model training; the competitor feature analysis model is trained based on the first target product and its corresponding training dataset, and the competitor feature analysis model automatically collects and analyzes competitor data based on the data of the second target product to obtain the competitor with the highest matching degree with the second target product and its feature identifier, thereby improving the comprehensiveness and accuracy of data processing, and thus improving the efficiency and accuracy of competitor analysis; an improved scheme for the second target product is generated based on the competitor and its feature identifier, thereby improving the design quality and competitiveness of the second target product.
[0011] Furthermore, the step of constructing a training dataset after feature identification of the competitor information includes: The competitor information is characterized from different dimensions, including competitor name, competitor type, user preference type, and competitor reference level.
[0012] As described above, by identifying the collected competitor information from different dimensions, it is possible to compare it with the competitor information searched by the competitor feature analysis model, quickly obtain more representative competitors and their feature identifiers, and break down competitor information from different dimensions to gain a more comprehensive understanding of competitor information, which is conducive to proposing targeted improvement solutions in the future.
[0013] Furthermore, step S2 includes: The relationship between the first target product and the first competitor information in the training dataset is trained, and the relationship between the first competitor information in the training dataset and its identifier is trained to obtain the training model; The training model is used to search for and identify the features of the second competitor information of the first target product. The second competitor information and its feature identifiers are compared and tested with the training dataset. When the accuracy of the test result does not reach the preset value, the training dataset is adjusted and the training model is retrained according to the adjusted training dataset until the accuracy of the test result reaches the preset value, thus obtaining the competitor feature analysis model.
[0014] As described above, by adjusting the training dataset according to the accuracy of the test results, the quality and representativeness of the training dataset are optimized, ensuring the dynamic updating and accuracy of the training dataset. Furthermore, by iteratively training the competitor feature analysis model, the stability and reliability of the competitor feature analysis model are improved. This enables the competitor feature analysis model to grasp the constantly updated competitor information and its functional modules, facilitating competitor analysis and improving the efficiency and accuracy of competitor analysis.
[0015] Furthermore, step S2 also includes: The second competitor information is sorted according to the competitor type and the competitor reference level, and the competitor with the highest matching degree with the first target product is identified from the second competitor information based on the sorting result.
[0016] As described above, by identifying competitor types and competitor reference levels, the competitor with the highest matching degree with the first target product can be identified, thus narrowing the scope of competitors, saving time and resources, avoiding wasting resources on a large number of irrelevant or low-matching competitors, and improving the efficiency of competitor analysis.
[0017] Furthermore, the step of generating an improved solution for the second target product based on the competitor and its feature identifiers includes: The gap between the competitor and the second target product is identified based on the feature identifier, and an improvement plan for the second target product is generated based on the gap.
[0018] As described above, by using feature identification to conduct a detailed analysis of competitors and the second target product, the gap between the two can be clearly understood. Based on this gap, an improvement plan for the second target product can be generated, and the design of the target product can be adjusted in a targeted manner to promote innovation in the target product, thereby improving the quality and competitiveness of the target product.
[0019] Please refer to Figure 2 Another embodiment of the present invention provides an automatic competitor analysis terminal, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the various steps of the above-described automatic competitor analysis method.
[0020] The automatic competitor analysis method and terminal described above are applicable to automatically collecting competitor information of target products and analyzing it according to different dimensions, thereby improving the efficiency and accuracy of competitor analysis. The following is a detailed description of specific implementation methods: Please refer to Figure 1 and Figure 3 Embodiment 1 of the present invention is: an automatic competitor analysis method, comprising the following steps: S1. Collect information on the first competitor of the first target product, and construct a training dataset after feature labeling the first competitor information.
[0021] In this embodiment, first competitor information of the first target product is collected. The first competitor information includes industry intelligence and competitor data. After feature identification of the first competitor information, a training dataset is constructed to lay the data foundation for subsequent training of the competitor feature analysis model and ensure the reliability of the competitor feature analysis model training. The preferred method for collecting the first competitor information is to automatically collect competitor data from multiple channels through web crawlers, API interfaces, and highly matched competitor retrieval and lightweight models. The collected data is then cleaned, deduplicated, and formatted to ensure the comprehensiveness, timeliness, and consistency of the data.
[0022] Furthermore, in this embodiment, constructing a training dataset after feature identification of the competitor information includes: feature identification of the competitor information from different dimensions, including competitor name, competitor type, user preference type, and competitor reference level. For example, taking game competitor analysis as an example, competitor types include shooting games, which are further classified according to gameplay, including: roguelike shooting, multiplayer cooperative shooting, RPG shooting, turn-based shooting, rhythm shooting, survival sandbox, loot shooting, tower defense shooting, pixel-style survival puzzle, vehicle shooting, tactical shooting, ARPG, MMORPG, MOBA-style survivor, RPG shooting, etc.; user preference types include sensory stimulation, which can be reflected in the player's actions in the game. During attack or defense actions, players can feel the impact of the action, accompanied by realistic blood splatter or object shattering effects. This can also be demonstrated when a large number of monsters appear during the game, allowing players to instantly kill them with headshots, creating an explosive visual effect that stimulates the player's adrenaline and brings a strong sense of instant gratification. The competitor reference level includes a rating system from one to five stars. For example, three stars are considered to have reference value, and five stars are considered to have the highest reference value. By identifying the features of the collected competitor information from different dimensions, it is possible to compare it with the competitor information found by the competitor feature analysis model, quickly obtain more representative competitors and their feature identifiers, and break down the competitor information from different dimensions to gain a more comprehensive understanding of the competitor information, which is conducive to proposing targeted improvement plans in the future.
[0023] S2. Train a competitor feature analysis model based on the first target product and its corresponding training dataset, including: The relationship between the first target product and the first competitor information in the training dataset is trained, and the relationship between the first competitor information in the training dataset and its identifier is trained to obtain the training model; The training model is used to search for and identify the features of the second competitor information for the first target product. The second competitor information and its feature identifiers are compared and tested with the training dataset. When the test result does not reach the preset value, the training dataset is adjusted, and the training model is retrained based on the adjusted training dataset until the accuracy of the test result reaches the preset value, thus obtaining the competitor feature analysis model.
[0024] In this embodiment, a training model is obtained by training the relationship between the first target product and the first competitor information in the training dataset, and training the relationship between the first competitor information in the training dataset and its identifier. This improves the training model's ability to search for and analyze competitors. The training dataset is adjusted accordingly based on the accuracy of the test results to optimize its quality and representativeness, ensuring dynamic updates and accuracy. Furthermore, the competitor feature analysis model is iteratively trained to improve its stability and reliability, enabling it to grasp continuously updated competitor information and its functional modules for competitor analysis, thus improving the efficiency and accuracy of competitor analysis. In addition, the model parameters can be continuously optimized through user annotations during the training process, improving the model's accuracy, adaptability, and predictive ability.
[0025] Furthermore, in this embodiment, step S2 further includes: sorting the second competitor information according to the competitor type and the competitor reference level, that is, competitors with higher reference levels will be recommended first among similar competitors. In addition, they can also be sorted according to the current popularity of similar competitors, such as annual revenue, popularity ranking order, etc. Based on the sorting results, the competitor with the highest matching degree with the first target product is identified from the second competitor information, narrowing the competitor range, saving time and efficiency, avoiding wasting resources on a large number of irrelevant or low-matching competitors, and improving the efficiency of competitor analysis.
[0026] S3. Input the data of the second target product into the competitor feature analysis model to obtain the competitor with the highest matching degree and its feature identifier, and generate an improvement plan for the second target product based on the competitor and its feature identifier.
[0027] In this embodiment, a competitor feature analysis model is used to automatically collect and analyze competitor data based on the data of the second target product, thereby obtaining the competitors with the highest matching degree with the second target product and their feature identifiers, improving the comprehensiveness and accuracy of data processing, and thus improving the efficiency and accuracy of competitor analysis; based on the competitors and their feature identifiers, an improvement plan for the second target product is generated, thereby improving the design quality and competitiveness of the second target product.
[0028] Furthermore, in this embodiment, the step of generating an improvement scheme for the second target product based on the competitor and its feature identifiers includes: identifying the gap between the competitor and the second target product based on the feature identifiers, generating an improvement scheme for the second target product based on the gap, and making targeted adjustments to the design of the target product to promote innovation in the target product, thereby improving the quality and competitiveness of the target product.
[0029] Please refer to Figure 2 Embodiment 2 of the present invention is: an automatic competitor analysis terminal 1, including a memory 2, a processor 3, and a computer program stored on the memory 2 and executable on the processor 3. When the processor 3 executes the computer program, it implements the various steps of the automatic competitor analysis method of Embodiment 1.
[0030] In summary, the present invention provides an automatic competitor analysis method and terminal. This method collects first competitor information for a first target product and identifies the competitor information from different dimensions, including competitor name, competitor type, user preference type, and competitor reference level. This information is then compared with competitor information searched by a competitor feature analysis model. A training dataset is constructed based on the identified competitor information, laying a data foundation for subsequent training of the competitor feature analysis model and ensuring the reliability of the model training. By training the relationship between the first target product and the first competitor information in the training dataset, and training the relationship between the first competitor information in the training dataset and its identifiers, a training model is obtained, improving the model's ability to search for and analyze competitors. The training model is then used to search for and identify second competitor information for the first target product. The second competitor information and its identifiers are compared and tested with the training dataset. If the test results do not reach a preset value, the training dataset is adjusted, and the training model is retrained based on the adjusted training dataset until the test result accuracy reaches a preset value. A competitor feature analysis model is obtained. Based on the accuracy of the test results, the training dataset is adjusted accordingly to optimize its quality and representativeness, ensuring dynamic updates and accuracy. The competitor feature analysis model is iteratively trained to improve its stability and reliability, enabling it to grasp continuously updated competitor information and their functional modules for effective competitor analysis, thus improving efficiency and accuracy. Secondary competitor information is sorted according to competitor type and reference level. Based on the sorting results, the competitor with the highest matching degree to the first target product is identified, narrowing the competitor range, saving time and improving efficiency. The data of the second target product is input into the competitor feature analysis model to obtain the competitor with the highest matching degree and its feature identifier. The gap between the competitor and the second target product is identified based on the feature identifier. An improvement plan for the second target product is generated based on the gap, allowing for targeted adjustments to the target product's design, promoting innovation, and ultimately improving the target product's quality and competitiveness.
[0031] Therefore, the embodiments of the present invention can achieve the following beneficial effects: By collecting competitor information through automated tools, constructing a training dataset after feature identification of the competitor information, and training a competitor feature analysis model based on the target product and the training dataset, the present invention enables automated data processing and analysis, significantly improving the efficiency and accuracy of competitor analysis. Compared with traditional methods, the present invention can acquire competitor data from multiple channels in real time, and perform in-depth analysis of the data using a trained competitor feature analysis model to obtain the competitors with the highest matching degree to the target product and their feature identifiers. Based on the feature identifiers, the present invention identifies the gap between the competitors and the second target product, and generates an improvement plan for the second target product based on the gap, narrowing the scope of competitors, significantly reducing manpower and time costs, and improving the comprehensiveness and accuracy of data processing. Simultaneously, it possesses self-learning and dynamic adjustment capabilities, enabling it to adjust analysis strategies based on the latest market data and competitor information, helping enterprises make faster and more accurate market decisions and enhance their competitive advantage.
[0032] The above description is merely an embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent modifications made based on the content of the present invention specification and drawings, or direct or indirect applications in related technical fields, are similarly included within the patent protection scope of the present invention.
Claims
1. An automated competitor analysis method, characterized in that, Including the following steps: S1. Collect information on the first competitor of the first target product, and construct a training dataset after feature labeling the first competitor information; S2. Train a competitor feature analysis model based on the first target product and its corresponding training dataset; S3. Input the data of the second target product into the competitor feature analysis model to obtain the competitor with the highest matching degree and its feature identifier, and generate an improvement plan for the second target product based on the competitor and its feature identifier.
2. The automatic competitor analysis method according to claim 1, characterized in that, The step of constructing a training dataset after feature identification of the competitor information includes: The competitor information is characterized from different dimensions, including competitor name, competitor type, user preference type, and competitor reference level.
3. The automatic competitor analysis method according to claim 2, characterized in that, Step S2 includes: The relationship between the first target product and the first competitor information in the training dataset is trained, and the relationship between the first competitor information in the training dataset and its identifier is trained to obtain the training model; The training model is used to search for and identify the features of the second competitor information of the first target product. The second competitor information and its feature identifiers are compared and tested with the training dataset. When the accuracy of the test result does not reach the preset value, the training dataset is adjusted and the training model is retrained according to the adjusted training dataset until the accuracy of the test result reaches the preset value, thus obtaining the competitor feature analysis model.
4. The automatic competitor analysis method according to claim 3, characterized in that, Step S2 further includes: The second competitor information is sorted according to the competitor type and the competitor reference level, and the competitor with the highest matching degree with the first target product is identified from the second competitor information based on the sorting result.
5. The automatic competitor analysis method according to claim 1, characterized in that, The improvement scheme for generating the second target product based on the competitors and their feature identifiers includes: The gap between the competitor and the second target product is identified based on the feature identifier, and an improvement plan for the second target product is generated based on the gap.
6. An automatic analysis terminal for competing products, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it performs the following steps: S1. Collect information on the first competitor of the first target product, and construct a training dataset after feature labeling the first competitor information; S2. Train a competitor feature analysis model based on the first target product and its corresponding training dataset; S3. Input the data of the second target product into the competitor feature analysis model to obtain the competitor with the highest matching degree and its feature identifier, and generate an improvement plan for the second target product based on the competitor and its feature identifier.
7. The automatic competitor analysis terminal according to claim 6, characterized in that, The step of constructing a training dataset after feature identification of the competitor information and constructing the training dataset based on the competitor data includes: The competitor information is characterized from different dimensions, including competitor name, competitor type, user preference type, and competitor reference level.
8. The automatic competitor analysis terminal according to claim 7, characterized in that, Step S2 includes: The relationship between the first target product and the first competitor information in the training dataset is trained, and the relationship between the first competitor information in the training dataset and its identifier is trained to obtain the training model; The training model is used to search for and identify the features of the second competitor information of the first target product. The second competitor information and its feature identifiers are compared and tested with the training dataset. When the accuracy of the test result does not reach the preset value, the training dataset is adjusted and the training model is retrained according to the adjusted training dataset until the accuracy of the test result reaches the preset value, thus obtaining the competitor feature analysis model.
9. The automatic competitor analysis terminal according to claim 8, characterized in that, Step S2 further includes: The second competitor information is sorted according to the competitor type and the competitor reference level, and the competitor with the highest matching degree with the first target product is identified from the second competitor information based on the sorting result.
10. The automatic competitor analysis terminal according to claim 6, characterized in that, The improvement scheme for generating the second target product based on the competitors and their feature identifiers includes: The gap between the competitor and the second target product is identified based on the feature identifier, and an improvement plan for the second target product is generated based on the gap.