Data reference method, and electronic device, storage medium and information processing apparatus

Abstract

A data reference method, comprising: step A0, providing a reference data set and a target data set; step S1, acquiring an attribute of the reference data set; and step A2, with reference to the attribute, evaluating a corresponding attribute of the target data set. By means of the method, the accuracy of data evaluation can be improved.

Description

Method for data reference, electronic device, storage medium and information processing apparatus TECHNICAL FIELD

[0001] The present application relates to the field of data processing, in particular to a method for data reference. BACKGROUND

[0002] At present, there is no effective method for evaluating data, which leads to insufficient evaluation of a data set that may have certain (common) attributes, and lack of automatic, statistically significant and accurate evaluation by machine, which makes users lack in-depth understanding of these attributes of the data set.

[0003] Taking the evaluation of patents in the patent industry as an example, the market value and technical value of a patent portfolio are usually evaluated by relying on the manual operation of industry experts and technical experts. Even if computer software is used for auxiliary processing, it is usually limited to the attributes of the patent portfolio itself, such as the total number of patents included in the patent portfolio, the number of patents according to the year, the composition of the technical classification number (such as IPC) involved in the patent portfolio, and the composition of the applicant involved in the patent portfolio. Such statistical data lack benchmarking with other data and lack a global perspective, which is a "local data" analysis of "local data". Such analysis has little significance for data evaluation.

[0004] In other scenarios, analysts may also compare the patents of two enterprises / schools, etc. Although such comparison constitutes a certain benchmarking, the results obtained are still limited to two "small data sets", both of which are "local" and do not have "global attributes". SUMMARY

[0005] Based on the above-mentioned defects, one of the purposes of the present application is to improve the accuracy of data evaluation.

[0006] To address one of the above or below technical problems, the inventors propose a concept of providing a reference data set for a target data set and configuring it on the user interface of the current application for the user to directly activate / call the software functions outside the current application during the operation of the current application.

[0007] According to this concept, in one embodiment of the present application, a method for data reference is proposed, which includes three steps:

[0008] A0. providing a reference data set and a target data set (such as a patent portfolio or a patent portfolio of a certain enterprise);

[0009] S1. Obtain attributes of the reference data set;

[0010] A2. Evaluate corresponding attributes of the target data set with reference to the attributes;

[0011] Wherein, the target data set and the reference data set have correspondence in the field and / or category to which they belong.

[0012] In addition, in an embodiment of the present application, a data reference method is also proposed, which includes three steps:

[0013] A0' provides a reference data set for the target data set (to be evaluated);

[0014] S1. Obtain attributes of the reference data set;

[0015] A2'. Present the target data set and the reference data set in association;

[0016] Wherein, the target data set and the reference data set have correspondence in the field and / or category to which they belong.

[0017] In addition, in an embodiment of the present application, a data reference method is also proposed, which includes three steps:

[0018] S7. Provide data of a first enterprise;

[0019] S2. Structurally present data of a first industry to which the first enterprise belongs;

[0020] S3. Present the data of the first enterprise with reference to the structured data of the first industry.

[0021] Optionally, the data reference method of some embodiments, wherein the method further includes the step of:

[0022] S4. Obtain industry attributes of the data of the first industry;

[0023] S5. Define / determine corresponding attributes of the data of the first enterprise with reference to the industry attributes.

[0024] Optionally, the data reference method of some embodiments, wherein the step S2 further includes the step S2-1A:

[0025] Cluster / classify the data of the first industry according to a first standard / first rule / first algorithm to obtain a plurality of reference data sets; and / or,

[0026] Cluster / classify the data of the first enterprise according to a first standard / first rule / first algorithm to obtain a plurality of target data sets; or,

[0027] The step S2 further comprises a step S2-1B:

[0028] The data of the first industry is clustered / classified to obtain a plurality of reference data sets; and / or,

[0029] The data of the first enterprise is clustered / classified to obtain a plurality of target data sets.

[0030] Optionally, the reference data sets have a larger number of elements relative to the target data sets; and / or,

[0031] Optionally, in some embodiments, in the corresponding field / category, the reference data sets have stronger statistical characteristics or representativeness relative to the target data sets; and / or,

[0032] Optionally, in some embodiments, in the corresponding field / category, the reference data sets have stronger statistical characteristics or representativeness relative to the target data sets.

[0033] Optionally, in some embodiments, the method further comprises a step A3:

[0034] The industry-level data set and the enterprise-level data set are respectively clustered / classified according to the same or substantially the same (classification) standard to obtain a plurality of reference data sets and a plurality of target data sets. The industry-level data set can be implemented as patent data associated with an industry (such as the lithium battery industry), and can be obtained by a patent navigation analysis project for the relevant industry.

[0035] Optionally, in some embodiments, the enterprise-level data set is a subset of the industry-level data set; and / or, the target data set is a subset of the reference data set.

[0036] Optionally, in some embodiments, the enterprise-level data set is a proper subset of the industry-level data set; and / or, the target data set is a proper subset of the reference data set.

[0037] Optionally, in some embodiments, the enterprise-level data set is the patent data of the first enterprise in the first category;

[0038] The first enterprise belongs to the first industry, and the industry-level data set is all or part of the patent data of the first industry in the first category.

[0039] Optionally, in some embodiments, the industry-level data set is patent data of top N (N = 1, 5, 10, 20, 30, 40, or 50, etc.) enterprises in the first category in terms of total number of patents; the top N enterprises all belong to the first industry, thereby realizing benchmarking of enterprise-level patent data; and / or,

[0040] The first category corresponds to a first product, thereby realizing benchmarking of product-level patent data between each enterprise.

[0041] Optionally, although in the method of some embodiments, the reference data set and the target data set are presented to the user in a correlated manner, for example, in step A2, the method of these embodiments can also include the following steps:

[0042] A8. Visually independently and / or separately present the reference data set and the target data set, so that the user visually distinguishes the reference data set and the target data set more clearly.

[0043] Optionally, the step A8 further includes the following sub-steps: present the plurality of reference data sets and the plurality of target data sets in a first list and a second list, respectively; the number of sets in the "plurality of reference data sets" is greater than the number of sets in the "plurality of target data sets"; and,

[0044] At least part of the target data sets in the second list correspond one-to-one to at least part of the reference data sets in the first list; the corresponding target data set and the reference data set are located in the same row.

[0045] Compared with the plurality of target data sets, the plurality of reference data sets have more categories / clusters

[0046] Optionally, the method of some embodiments further includes step A7: arranging the corresponding target data set and the reference data set in the same row.

[0047] Optionally, in some embodiments, compared with the plurality of target data sets, the plurality of reference data sets have more categories / clusters; further optionally, the number of sets in the plurality of reference data sets is greater than the number of sets in the plurality of target data sets;

[0048] The number of the reference data sets is more than the number of the target data sets. That is, the reference data sets include two parts: the first part data sets and the second part data sets. The data sets in the first part data sets respectively correspond to the data sets in the target data sets in terms of classification / clustering. The second part data sets are the data sets that are more than the target data sets (or more data sets).

[0049] The first part data sets and the second part data sets are obtained by clustering / classifying the industry-level data sets and the enterprise-level data sets respectively according to the same standard in step A4. Therefore, the classification / clustering of the more data sets (i.e., the second part data sets) is different from the classification / clustering of the other data sets (i.e., the first part data sets) in the reference data sets, but the first part data sets and the second part data sets are both differentiated from the industry-level data sets, and the classification / clustering of the first part data sets and the second part data sets are related to each other to some extent.

[0050] The first part data sets in the reference data sets correspond to the target data sets in terms of classification / clustering. That is, the classification / clustering of each data set in the first part data sets corresponds to the classification / clustering of one of the target data sets, and the classification / clustering of each data set in the target data sets corresponds to the classification / clustering of one of the first part data sets. Further, each data set in the first part data sets can only find a unique data set in the target data sets that has a corresponding relationship in terms of classification / clustering, and vice versa. Therefore, the second part data sets are different from the target data sets in terms of classification / clustering, but are related to each other.

[0051] In another aspect, the data sets in the first part data sets respectively correspond to the data sets in the target data sets in terms of classification / clustering.

[0052] Optionally, the method of some embodiments further includes the following steps:

[0053] A5 arranging the reference data sets in the first list according to the similarity of classification / clustering;

[0054] A6 will be closer to the classification / clustering of the data sets, more closely arranged in the first list.

[0055] This means: will be more similar, closer to the classification / clustering of the data sets, arranged in the first list of more closely. For the similarity, similarity, proximity between these data sets, the measurement can be obtained by means of the Euclidean distance, cosine similarity, etc.

[0056] Optionally, the reference data set is a reference patent data set, and the target data set is a target patent data set.

[0057] Optionally, in the method of some embodiments, the attributes of the reference patent data set include at least one of:

[0058] Market value, or technical value, legal value, interdisciplinary degree.

[0059] Optionally, in the method of some embodiments, the attribute of the reference patent data set is a trend feature; the trend feature is a market trend degree, at least partially representing / reflecting the market value of the technical category corresponding to the reference patent data set.

[0060] The trend feature is a technical trend degree, at least partially representing the technical value of the patent set.

[0061] Optionally, the method of some embodiments or step A2 therein further comprises the following steps:

[0062] According to the descending order / ascending order of the evaluation index of the reference data set, the plurality of reference data sets are arranged in the first list. This is conducive to efficient browsing according to the weight of the index.

[0063] Statistical analysis is performed on the plurality of reference data sets to obtain the evaluation index.

[0064] Generally, technological trends are driven and dominated by large companies (e.g., the top N companies in an industry). These large companies define their technological paths through a large number of patent applications. Therefore, the patent data of large companies, especially those ranking among the top in patent applications, can largely reflect technological trends. Furthermore, a trend can be reflected in the technological paths chosen by many companies in the industry, or the R&D directions chosen by a majority (major) of companies. Therefore, analyzing the incremental and existing patent data of a particular company or industry can reveal these trends to a significant extent. Especially through horizontal comparisons between different clusters within the industry, it is possible to identify those technological paths chosen by more companies and with a higher chance of success.

[0065] Optionally, the number of patent applications in some embodiments includes one or more of the following: invention applications, invention patents, design patents, and utility model patents. Preferably, it is the sum of the number of invention applications and utility model patents, or the sum of the number of invention applications, utility model patents, and design patents.

[0066] Optionally, some embodiments of the method further include the step of: assigning a corresponding evaluation index to the target patent data set based on the evaluation index of the reference patent data set and according to the correlation / similarity between the target patent data set and the reference patent data set.

[0067] Optionally, in some embodiments of the method, similarity can be measured by the following numerical values:

[0068] i) The proportion of the intersection of the target patent dataset and ii) the reference patent dataset within the target patent dataset; or,

[0069] iii) The proportion of the intersection of the target patent dataset and the reference patent dataset in the reference patent dataset; or,

[0070] a) The ratio of the size of the intersection of the target patent data set and the reference patent data set to b) the size of the union of the target patent data set and the reference patent data set.

[0071] In another embodiment of this application, a method for data reference or comparison is proposed, including the following steps:

[0072] SB1) provides a first and a second set of data that are comparable / correlated;

[0073] SB2) Cluster / classify the first data set and the second data set respectively using the same or similar standards / rules to obtain a first group of multiple classifications and a second group of multiple classifications;

[0074] SB3) presenting / associating presenting the first part of categories in the first group of multiple categories with the first part of categories in the second group of multiple categories; and / or, presenting / associating presenting the second part of categories in the first group of multiple categories with the second part of categories in the second group of multiple categories;

[0075] Optionally, the step SB3 further comprises: presenting / associating presenting i) the data of the first part of categories in the first group of multiple categories with ii) the data of the first part of categories in the second group of multiple categories; and / or, presenting / associating presenting i) the data of the second part of categories in the first group of multiple categories with ii) the data of the second part of categories in the second group of multiple categories.

[0076] Optionally, from the perspective of set, the first group of multiple categories A and the second group of multiple categories B can be a containing relationship or an equal relationship, or an unequal relationship, or a non-containing relationship, for example or

[0077] Preferably, in some embodiments, the first group of multiple categories A and the second group of multiple categories B are and the first group of multiple categories A and the second group of multiple categories B have an intersection, which is the first part of categories. In other words, the first group of multiple categories and the second group of multiple categories both include the first part of categories, and the second part of categories is: i) the categories that belong to the first group of multiple categories but do not belong to the second group of multiple categories, and ii) the categories that belong to the second group of multiple categories but do not belong to the first group of multiple categories.

[0078] Optionally, the first data set is data of a first industry, and the second data set is data of a second industry. Alternatively, the first data set is data of a first enterprise, and the second data set is data of a second enterprise.

[0079] In some embodiments, the first data set is data of a first industry, and the second data set is data of a first enterprise. Wherein the data of the first group of multiple categories can be a plurality of reference data sets, and the data of the second group of multiple categories can be a plurality of target data sets.

[0080] In addition, the processing steps in the methods of any other embodiments for processing data sets such as enterprise-level data are also applicable to the processing of industry-level data in other embodiments, for example: the "enterprise-level data" in the steps of any embodiment of the present application, if any, can be replaced by "industry-level data (set)", "first / second / third industry data (set)", or words with the same meaning, thereby forming a new step for processing "industry-level data", and combining with other steps (for processing enterprise-level / industry-level data) to form a method for processing "industry-level data" in a variant.

[0081] Similarly, the processing steps in the methods of any other embodiments for processing data sets such as industry-level data are also applicable to the processing of enterprise-level data in other embodiments, for example: the "industry-level data" in the steps of any embodiment of the present application, if any, can be replaced by "enterprise-level data (set)", "first / second / third enterprise data (set)", or words with the same meaning, thereby forming a new step for processing "enterprise-level data", and combining with other steps (for processing enterprise-level / industry-level data) to form a method for processing "enterprise-level data" in a variant.

[0082] Data refers to the collection, analysis and interpretation of digital information. In the process of using data, it is necessary to show the overall situation, rather than isolating part of the data from the big background and looking at it in isolation. The latter can consolidate the shaky data evaluation by distorting the data, but this method will make the relevant points and data evaluation become blind and reduce the accuracy, and the evaluation conclusion is also questionable.

[0083] In some embodiments, a reference data set is provided for the target data set to be evaluated, and the corresponding attributes of the target data set are evaluated with reference to the attributes of the reference data set; this is equivalent to evaluating the corresponding attributes of the target data set against the attributes of the reference data set as a background, and the target data set and the reference data set have correspondence in their respective fields and / or categories, i.e. comparability between them. By referring to the reference data set and its attributes as a reference background / reference object, the evaluation of the data can be more comprehensive and accurate, avoiding the blindness of independent evaluation and separate assessment of the target data set, and also improving the accuracy of data evaluation and reducing the difficulty of identifying trend information in the data.

[0084] In other embodiments, the target data set and the reference data set are presented in association. Even if it is not necessary to provide a direct evaluation of the target data set or assign evaluation indicators through a computer, users can still obtain estimates and evaluations of various indicators of the target data set by referring to the attributes of the reference object when browsing the two data sets presented in association. For example, suppose that after clustering an industry-level patent data set, a large cluster A is generated. This reflects a confirmed technological trend in the industry in the corresponding technical direction. If the company also has a corresponding patent data cluster A' in this technical direction, or if the number of data elements in cluster A and cluster A' is comparable in terms of annual growth rate, then the company has basically followed at least part of the market trend, implemented at least part of the correct R&D activities, and followed at least part of the correct technical route in the industry.

[0085] In some embodiments, the target dataset (for example, an enterprise-level dataset) is a proper subset of the reference dataset (for example, an industry-level dataset). Because the reference dataset contains more data than the target dataset (to be evaluated), the analysis of the reference dataset and the resulting metrics are more statistically significant. Due to the smaller size and lack of statistical significance of the target dataset, some industry-level trends are difficult to identify through statistical analysis of limited enterprise-level data.

[0086] Furthermore, the first part of the industry-level dataset directly corresponds to the enterprise-level dataset. The "second part" of the dataset, which represents the difference between the two, reflects their different components. This "second part" not only has a certain correlation with the first part of the industry-level dataset, but also, in terms of clustering / classification, it can link to more data—including peripheral industry data—that the enterprise-level dataset lacks. This data also has direct / indirect connections with the enterprise-level dataset, representing a more macro-level relationship, and is not limited to any single individual company or its direct competitors. Therefore, the industry-level dataset, including the "second part," can, in turn, serve as a more macro-level reference, providing richer dimensions for evaluating individual enterprise datasets. Attached Figure Description

[0087] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0088] FIG. 1a is a flowchart of a data referencing method according to one embodiment of the present application;

[0089] FIG. 1b is a diagram showing the relative proportion of a reference data set and a target data set in one embodiment of the present application;

[0090] FIG. 2 is a diagram showing a user interface of a data referencing method implemented according to an embodiment of the present application;

[0091] In the description of the drawings, identical, similar or corresponding reference numerals represent identical, similar or corresponding elements, elements or functions. DETAILED DESCRIPTION

[0092] The terminology used in the description herein is for the purpose of describing particular embodiments only and is not intended to be limiting. As used in the description of the various embodiments and the appended claims herein, the singular forms "a", "an" and "the" are intended to include plural forms as well, unless the context clearly indicates otherwise. It will also be understood that the term "and / or" as used herein refers to and encompasses any and all possible combinations of one or more of the associated listed items. It will be further understood that the terms "comprises" and / or "comprising," when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof.

[0093] The word "by" as used herein, depending at context, can be interpreted to mean "by way of," "by means of" or "by virtue of." The word "if' as used herein, depending at context, can be interpreted to mean "when" or "responsive to the determination" or "in response to detecting," among others. Similarly, "when" or "when a" as used in some embodiments, can be interpreted to mean "if" or "in response to the determination" or "in response to detecting," among others. Similarly, the phrase "if (a stated condition or event) occurs," as used herein, can be interpreted to mean "when (a stated condition or event) occurs" or "in response to determining" or "in response to detecting (a stated condition or event) unless specified otherwise.

[0094] It should be understood that, although the terms first, second, third, etc. can be employed in this disclosure to describe various information, the information should not be limited to these terms. These terms are only used to differentiate one piece of information from another. For example, without departing from the scope of the disclosure, first can also be referred to as second, and vice versa. Depending on the context, the word "if' as used herein can be interpreted as "when" or "upon" or "in response to determining".

[0095] The present application is further illustrated below by way of examples, but the present application is not limited to the examples.

[0096] Market value, technical value, etc. often have global attributes, are a kind of trend reflecting attributes, and have macro significance. Therefore, market value, technical value, etc. need to be measured based on statistical calculation, and it is difficult to calculate based on data of one or a few companies (such as market data, patent portfolio, etc.).

[0097] Similarly, data information of a single company or two companies is also micro data, which lacks statistical significance. Simple benchmarking of data information between two individual companies is also blind, because the technologies of the two companies may deviate from the "industry track" or even be wrong, which makes the "strong / weak" and "high / low" conclusion between the two companies meaningless and even misleading.

[0098] Therefore, the inventors propose that the evaluation of an individual company or the benchmarking of two individual companies is based on a same type of larger relevant data set as a reference background for "evaluation of a single company" or "comparison between multiple companies". Under the macro reference background, the analysis and comparison of micro information are more accurate, meaningful, and minimize deviation / error.

[0099] In view of the above concept, in one embodiment of the present application, a data reference method is proposed, which includes steps S7, S2, and S3.

[0100] In step S7, enterprise-level data Y1 and corresponding industry-level data X1 (enterprise-level data Y1 and industry-level data X1, which can also be referred to as enterprise-level data and industry-level data, or enterprise-level data set and industry-level data set in other embodiments) are provided;

[0101] In step S2, the industry-level data X1 is structured and presented;

[0102] In step S3, the enterprise-level data Y1 is presented against the structured industry-level data X1. For example, the first industry-level data is taken as a reference structure, and the data of the first enterprise is arranged according to the arrangement structure in the reference structure, i.e., the data of the first enterprise is arranged according to the structure of the reference structure, which makes the presentation of the enterprise-level data Y1 to the user ordered, and the order is obtained by presenting the enterprise-level data Y1 against the industry-level data X1. To some extent, the enterprise-level data Y1 also has a structure corresponding to the reference structure. In this way, the enterprise-level data Y1 is presented in a certain degree of decomposition and structuring against the background of the industry-level data X1, so the enterprise-level data Y1 has a basic same / similar structure as the industry-level data X1, and has stronger readability, orderliness and guiding significance to the user.

[0103] Optionally, in the data reference method of some embodiments, steps S4 and S5 are further included. In step S4, the attributes of the industry-level data X1 are obtained; and since the industry-level data X1 has been structured, the attribute calculation of the structured data X1 can actually be the attribute calculation of each part of the data after structuring, which has higher accuracy.

[0104] In step S5, the corresponding attributes of the enterprise-level data Y1 are determined against the above attributes.

[0105] Alternatively, in step S5, the corresponding attributes of the enterprise-level data Y1 are defined against the above attributes. This means that the calculation of the attributes of the enterprise-level data Y1 does not completely depend on its own characteristics, and the attributes of the industry-level data X1 also have a certain weight in the calculation of the corresponding attributes of the enterprise-level data Y1. This is also a way of referring to the industry-level data X1 in the evaluation of the enterprise-level data Y1.

[0106] For example, when the industry-level data X1 and the enterprise-level data Y1 are clustered to generate the multiple reference data sets A1, A1', B1, C1 and the multiple target data sets a1, b1, c1, respectively, first, in the data reference method of this embodiment, the evaluation index of the "technology activity" of the first reference data set A1 can be calculated independently, and the evaluation index of the "technology activity" of the first target data set a1 can be calculated independently. For example, the "technology activity" of the first target data set a1 is calculated completely by using some data characteristics (such as the increment, the inventory, etc. in years) of the first target data set a1 itself. Then, step S5 further includes a sub-step of: according to the evaluation index of the "technology activity" of the first reference data set A1 and the evaluation index of the "technology activity" of the first target data set a1, weighting determines the final target evaluation index of the first target data set a1: the technology activity. This means that the technology activity of the first target data set a1 depends not only on the number of data elements, the change, etc. of its own, but also on the value of the "technology activity" of the first reference data set A1 as the reference background of the first target data set a1.

[0107] Those skilled in the art should understand that when calculating the value index of the "technology activity" and the "technology advancement" of the first reference data set A1, various methods for calculating the corresponding value index of the "single patent" in the prior art can also be used, for example: first, the value calculation tool published by the patentics company is used to calculate the value index such as the "stability" of each "single patent" in the first reference data set A1. Then, the value index such as the "stability" of each patent in the first reference data set A1 is averaged or weighted to obtain the value index such as the "stability" of the first reference data set A1 as a whole.

[0108] The "stability" described above is an index for evaluating the stability of a patent, which is obtained by a specific calculation formula. The lower the value, the larger the protection range of the patent and the stronger the innovation ability, which can reflect the technology advancement of the patent to a certain extent.

[0109] Optionally, in the data reference method of some embodiments, step S2 further includes the following sub-steps of clustering / classification / technology breakdown. For example, step S2-1A: according to the first standard / first rule / first algorithm, the industry-level data X1 is clustered / classified / broken down / technology broken down to obtain one or more reference data sets A1, A1', B1, C1, which are located on the multiple lines LA1, LA2, LB3, LC4, as shown in FIG. 2.

[0110] In addition, according to the first criterion / first rule / first algorithm, the enterprise-level data Y1 is clustered / classified / decomposed / technically decomposed to obtain one or more target data sets a1, b1, c1 located on multiple rows LA1, LB3, LC4, as shown in FIG. 2.

[0111] Of course, it can be understood that the clustering / classification / decomposition / technical decomposition is performed by using the same criterion, which does not mean that the specific classification / clustering algorithm is completely consistent, and details are omitted here.

[0112] Step S2 can further include a sub-step S2-1B:

[0113] The industry-level data X1 is clustered / classified / decomposed / technically decomposed to obtain one or more reference data sets A1, A1', B1, C1 located on multiple rows LA1, LA2, LB3, LC4, as shown in FIG. 2; and / or,

[0114] The enterprise-level data Y1 is clustered / classified / decomposed / technically decomposed to obtain one or more target data sets a1, b1, c1 located on multiple rows LA1, LB3, LC4.

[0115] For another example, step S2 can further include a sub-step S2-1C:

[0116] The first product (for example, lithium iron phosphate battery) related data in the industry-level data X1 is clustered / classified / decomposed / technically decomposed to obtain one or more reference data sets A1, A1', B1, C1; wherein the industry-level data X1 can be, for example, patent data of the lithium battery industry.

[0117] The first product related data in the enterprise-level data Y1 is clustered / classified / decomposed / technically decomposed to obtain one or more target data sets a1, b1, c1. Wherein the enterprise-level data Y1 can be, for example, patent data of BYD Company, and the first product can be the lithium iron phosphate battery of BYD Company.

[0118] Wherein, the clustering processing can be based on the same criterion / rule / algorithm; or the classification processing can be based on the same criterion / rule / algorithm; or the decomposition / technical decomposition processing can be based on the same criterion / rule / algorithm. The multiple target data sets a1, b1, c1 generated by the clustering, classification, (technical) decomposition and other processing steps are multiple different data sets distinguished according to the functional categories. Correspondingly, the multiple reference data sets A1, A1', B1, C1 are also different data sets distinguished according to the functional categories.

[0119] Alternatively, in some alternatives, the plurality of target data sets a1, b1, c1 are a plurality of different data sets differentiated by product category. Correspondingly, the plurality of reference data sets A1, A1', B1, C1 are also different data sets differentiated by product category.

[0120] In other alternatives, the plurality of target data sets a1, b1, c1 are a plurality of different data sets differentiated by application category. Correspondingly, the plurality of reference data sets A1, A1', B1, C1 are also different data sets differentiated by application category.

[0121] Further, since both the industry-level data set X1 and the enterprise-level data set Y1 are clustered / classified by the same criteria (or rules, algorithms), the plurality of target data sets a1, b1, c1, which are generated, correspond to a portion of the plurality of reference data sets (if the number of clustered reference data sets is more than the number of target data sets), or all of the plurality of reference data sets (if the number of clustered reference data sets is equal to the number of target data sets), in terms of classification / category.

[0122] For example, in sub-step S2-1D of step S2 of the data referencing method of some embodiments: the industry-level data X1 is clustered / classified / broken down / technically broken down with “product” as the clustering unit to obtain one or more reference data sets A1, A1', B1, C1. Alternatively, the industry-level data X1 is clustered / classified / broken down / technically broken down with “product” as the granularity of clustering to obtain one or more reference data sets A1, A1', B1, C1.

[0123] Further, i) the plurality of target data sets a1, b1, c1 and ii) the plurality of reference data sets A1, B1, C1 and their attributes, which correspond in terms of field and / or category, are presented in visual association, as shown in FIG. 2, so as to facilitate the user to visually compare i) the plurality of target data sets a1, b1, c1 and ii) the plurality of reference data sets A1, B1, C1, and the corresponding attributes of both.

[0124] Each of these reference data sets corresponds to a type of product in the industry (for example, ternary lithium batteries, lithium iron phosphate batteries, etc.). In this way, the reference data set related to a new product in the industry obtained by clustering can be used as a reference background to present the correlation between the target data set and the reference data set, and it is easy to present the information A1' of one or more new products in the industry. Since the product type corresponding to the reference data set A1' belongs to the lithium battery industry and is not owned by the Ningde Times Company, through the correlation presentation mode of the list in FIG. 2, the enterprise user can easily find the information of these new products. Of course, the new product information here is the "new" product information that the enterprise user (for example, Ningde Times) corresponding to the target data set has not developed or applied for a patent, and this "new" product itself may have existed in the lithium battery industry for some time, and is not a completely new type of product in the lithium battery industry in the past year or half a year.

[0125] Preferably, through the clustering / classification / (technical) decomposition steps of structuring the enterprise-level data Y1 and the industry-level data X1, when the enterprise-level data Y1 and the industry-level data X1 have (in category) corresponding clustering / classification / (technical) decomposition, respectively, the enterprise-level data Y1 or each clustering / classification / (technical) decomposition module thereof can be arranged in the sub-step of step S3, with reference to the data arrangement structure of the structured industry-level data X1.

[0126] In some alternative embodiments, in the sub-step of step S3, the enterprise-level data Y1 or each clustering / classification / (technical) decomposition module thereof can also be arranged with reference to the data arrangement (for example, the first column γ1 in FIG. 2) of the structured industry-level data X1, as shown in the second column β2 in FIG. 2. This means that the enterprise-level data Y1 or each clustering / classification / (technical) decomposition module thereof is distributed and presented with reference to the structured industry-level data X1 or the data arrangement thereof (for example, the first column γ1 in FIG. 2), as shown in the second column β2 in FIG. 2. Or, the enterprise-level data Y1 or each clustering / classification / (technical) decomposition module thereof is distributed and presented at least partially corresponding to the data arrangement structure (in category) of the industry-level data X1. Wherein, the industry-level data X1 and the enterprise-level data Y1 can be arranged in the display of the electronic device or the table presented thereon.

[0127] Of course, preferably, in most embodiments, the industry-level data X1 is only the patent data of the past M years (M = 2, 3, 5, 8, 10, etc.) in the union of i) all enterprises or ii) part of the main / important enterprises in the relevant industry, for example, as of June 30, 2024, only the patent data in the above-mentioned patent data union whose filing date is within the interval [January 1, 2021, December 31, 2022] is counted as the industry-level (patent) data. This way of defining industry-level data takes into account the fact that patent data will not be disclosed until 18 months after the filing date. In fact, especially in China, many (invention) patents are disclosed in advance, and utility model patents are even more so. Therefore, the above-mentioned time interval can also be defined as [January 1, 2022, December 31, 2023] or a more recent time period, such as [January 1, 2022, June 30, 2024], which is also meaningful in terms of statistical significance.

[0128] The reason for defining enterprise-level data in this way is that in some patent layout scenarios, only the distribution trend of industry-level patent data in the past 2-5 years is needed to guide and guide the layout of individual enterprises A. If the industry-level data X1 includes patent data from a long time ago, those calculations of data statistics from a long time ago (e.g., 50 years ago) will have a negative impact and mislead the current patent layout, because technology is developing rapidly, and the closer it is to the current year, the faster it develops, and the more (patent) data it generates, the more guidance it provides. Therefore, the correlation between long-term R&D data (such as patent data) and current R&D activities may be small and lack guidance.

[0129] Similarly, the same applies to enterprise-level data X1, for example, in some embodiments, the industry-level data in the past three years can be structured, and then the structured industry-level data X1 is referred to to present the enterprise-level data X1. Or, the industry-level data in the past three years is structured by clustering, classification, (technical) decomposition, etc., and then the corresponding clustering, classification attributes in the enterprise-level data X1 are evaluated and assessed with reference to the clustering, classification attributes in the past three years.

[0130] Optionally, in the time dimension, the industry-level data X1 is only the patent data in the time interval from the previous Pth year to the Qth year in the union of i) all enterprises or ii) part of the main / important enterprises in the relevant industry, where Q>P, Q-P>1, P=1, 0.5, 2, or 1.5, etc., for example, P=1.5, 2, 3, 5, Q=2.5, 3, 3.5, 4, 4.5, 5, 5.5, etc., or P=2.5, 3, 4, 5, Q=3.5, 4, 5.5, 6, etc.

[0131] Optionally, in the geographical dimension, the industry-level data X1 can be limited to the data of one or more countries (e.g., multiple countries with similar social needs, economic systems, social culture, and public behavior habits), or regions, i.e., these industry-level data are accepted by the patent office of one or more countries and applied for and examined through the country / countries. In this way, the calculation of the "trend" and other attributes of the industry-level data will also be for the specific country / countries, and if the company-level data Y1 is benchmarked or laid out against such industry-level data X1, the accuracy of cross-border layout and layout for a specific country is improved. This is because the systems, laws, and market operations of each country are different, so even if a patent portfolio has high market value in country A, it may not necessarily have high market value in country B. Therefore, benchmarking and laying out enterprise-level data Y1 against industry-level data belonging to different countries is more targeted and meaningful in the application scenario of international patent layout.

[0132] In other embodiments, the first column and the second column in FIG. 2 can respectively arrange two industry-level data. For example, the first industry-level data and the second industry-level data are respectively structured and structured presented. This structuring (presentation) can be implemented by means such as step A4 in other embodiments, for example: clustering / classifying the first industry-level data and the second industry-level data respectively by the same standard / rule / algorithm to obtain a first group of classifications / clusters (e.g., a plurality of reference data sets) and a second group of classifications / clusters (e.g., a plurality of target data sets). Then, a) the first group of classifications / clusters (in the first column) obtained from the first industry-level data and b) the second group of classifications / clusters (in the second column) obtained from the second industry-level data are compared and benchmarked.

[0133] For two comparable industries, such as lithium battery cars and fuel cars, and their (patent) data, some categories (or clusters) in the first set of categories and the second set of categories are repeated, corresponding, such as tires, steering wheels, etc., while some other categories (or clusters) are misplaced, such as no corresponding each other in the first column and the second column, such as the "engine" category in the second set of categories for fuel cars, the "electric motor" category in the first set of categories, etc. Preferably, according to the calculated values of the respective market value and other value indexes of these misplaced categories, the overall market value, market / technology development potential, and development trend of the two industries can be basically judged. For the above-mentioned categories / clusters of tires, steering wheels, etc. that have corresponding relationships in the data of the two industries, there are generally similar value indexes, because in the real two industries, the related technologies of these tires, steering wheels, etc. components can be reusable.

[0134] Of course, in some steps of these methods, various value indexes of the industry-level data can be calculated based only on the value indexes of the "misplaced" (or complementary, not corresponding to each other, no corresponding to each other) "non-common" technical categories (such as electric motor, engine, etc.) in the second set of categories and the first set of categories. In other alternative embodiments, various value indexes of the relevant industry-level data can also be calculated based on the common technical categories (such as tires, steering wheels, etc.) in the above-mentioned second set of categories (or clusters) and the first set of categories (or clusters), especially in the case that the various value indexes calculated in the above-mentioned second set of categories (or clusters) and the first set of categories (or clusters) of tires, steering wheels, etc. categories do not tend to be the same.

[0135] It should be noted that in the above process of benchmarking two different industries such as lithium battery cars and fuel cars, the first part of the categories in the first set of categories and the second set of categories are corresponding, similar, or the same, while the second part of the other categories are different, mutually different, and not corresponding to each other. In some embodiments of the method, a step can be included to distribute the data of the above-mentioned first part of categories in the first set of categories and the second set of categories in the first list and the second list, and to present (or simultaneously present the value index of each of the first part of categories) corresponding / associated or to compare the value indexes of the data of the first part of categories in the first set of categories and the second set of categories. The comparison between such similar / same categories can more clearly reflect the development trend of the two different industries to the user.

[0136] Optionally, in the method of some other embodiments, the step of distributing, correlatively presenting (or re-presenting the value indicator of each of the second group of categories), or making a whole comparison of the value indicators between the data of the second group of categories in the first group of categories and the second group of categories, i.e. not necessarily presenting, correlatively presenting one-to-one, the data of the second group of categories in i) the first group of categories and ii) the second group of categories in the category corresponding to each other, can be included. In fact, the second group of categories in the first group of categories and ii) the second group of categories in the second group of categories also have no similarity and correspondence. Because the two industries are different because the core links of the industries and the core components of the products have differences, for example: the motor and lithium battery pack in the lithium battery car, and the engine and exhaust filter in the fuel car, etc. The comparison of the value indicators between the (patent) data corresponding to these different and distinctive parts can exactly reflect the strength of the core competitiveness of the two industries / industries, the advancement of the core technology, and the market value. For example, the "value indicator" calculated by the increase of the number of patent applications, the storage amount, etc. reflects the market value of the motor and lithium battery pack in the lithium battery car showing a trend of increasing year by year, while the market value of the engine and exhaust filter in the fuel car shows a trend of decreasing year by year, and after normalization calculation, the market value of the former is higher than that of the latter in the whole, which reflects that the development level of the lithium battery car industry is higher than that of the fuel car industry in the market to a certain extent through this comparison step.

[0137] Furthermore, since the embodiments are aimed at the industry-level data of two comparable industries, the various value indices calculated based on the first group of categories and the second group of categories are calculated based on the large amount of industry-level data, so that the large amount of data and the statistical significance thereof make these value indices have high accuracy. Therefore, the growth potential and market value of the two industries or their respective technical routes can be calculated by comparing the two different but possibly competitive industries.

[0138] It should be noted that all the above steps of decomposition and comparison of the data sets of the industry level of two different but comparable industries (or industries) are also applicable to the decomposition and comparison of the data sets of the enterprise level of two different entities (such as enterprises and institutions, etc.), for example, replacing the "first industry level data" and "second industry level data" in some steps in the related embodiments with "first enterprise level data" and "second enterprise level data" respectively, thereby forming steps for enterprise level data. Therefore, it is not necessary to repeat them here.

[0139] The discovery of information that exists in the industry but is not involved in the user's own research and development activities, the patent layout of the user, or the benchmarking of the technology roadmap, the identification of new technical elements in the industry, has a strong research and development guidance significance. Especially those reference data sets A1', which are neither the same as nor related to the target data sets a1 owned by individual enterprises. This information can prompt the enterprise user: whether it is necessary to extend its research and development activities to the (new) application direction, (new) technical subdivision direction, (new) product direction corresponding to the reference data set A1'. For example, in the new energy industry, if the (patent) data in the industry is often compared with the patent benchmarking and comparison through the method of the embodiments of the present application, then in a certain period of time, it can help individual enterprises such as Ningde Times in the lithium battery industry: discover these new competitive technology routes, such as solid-state batteries, nanometer power generation, etc., and discover related (potential) competitors at the stage of other new technology routes.

[0140] Moreover, for the first reference data set A1 corresponding to the first target data set a1 in the field and / or category, in the calculation of the market value and other indicators of the first target data set a1, it does not completely rely on some data characteristics of the first target data set a1 itself, but the corresponding "market value" and other evaluation indicators of the first reference data set A1 will also have a certain weight in the calculation of these attributes of the first target data set a1. The field / category of the reference data set A1 and the target data set a1 is corresponding, same or similar. Relative to the target data set a1, the reference data set A1 has a larger number of elements and stronger statistical characteristics or representativeness in the subdivision of the research and development field and the product development direction. For example, "market value" can also be understood as whether it meets the market (demand) trend to some extent, and "trend" is the choice of product development and technology route by most enterprises in the related industry in response to the demand of a large number of users on the market. Since patent data can also reflect the direction of research and development, the patent data of these "most enterprises" can more accurately reflect the demand trend of the market, the direction of research and development investment in the industry, and the market value of related technologies than the (patent) data of a single enterprise. Therefore, here, the attributes and parameters of the reference data set A1 are used as influencing factors / factors in the calculation of the corresponding attributes of the target data set a1, which is meaningful for the final improvement of the accuracy of the market value and technical value and other evaluation indicators of the target data set a1. For example, first, the evaluation indicators of a) the reference data set A1 and b) the target data set a1 are calculated independently, and then the two are weighted and summed to calculate the value of the target evaluation indicator of the target data set a1. Or, directly assign the market value evaluation indicator of the reference data set A1 to the target data set a1 as its market value indicator.

[0141] Further, in one embodiment of the present application, a data reference method is also proposed, as shown in FIG. 1a, including steps A0, S1 and A2.

[0142] In step A0, the reference data sets A1, B1, C1 and the target data sets a1, b1, c1 are provided, for example, one or more patent pools / patent portfolios / patent sets of an enterprise, or a single enterprise patent portfolio / set;

[0143] In step S1, the attributes of the reference data sets A1, B1, C1 are obtained;

[0144] In step A2, the corresponding attributes of the target data sets a1, b1, c1 are evaluated by referring to the attributes of the reference data sets A1, B1, C1 described above;

[0145] Optionally, the target data sets a1, b1, c1 and the reference data sets A1, B1, C1 have correspondence in the (technical) field and / or category to which they belong.

[0146] Of course, in step A2, the reference to the corresponding attributes of the reference data sets A1, B1, C1 in the evaluation of the relevant attributes of the target data sets a1, b1, c1 is not limited to the synchronous presentation of the two data sets, e.g. on the display of the electronic device. As shown in FIG. 2, the synchronous presentation of the two data sets can formally enhance the reference effect between the two for the user, but in fact, the attributes of the reference data sets A1, B1, C1 can be only used in the background to calculate the corresponding attributes of the target data sets a1, b1, c1, while only the relevant attributes of the target data sets a1, b1, c1 are presented on the visual interaction layer of the display.

[0147] In addition, in an embodiment of the present application, a data reference method is also proposed, which includes three steps:

[0148] A0’ provides a reference data set A1, B1, C1 for the target data set a1, b1, c1 (to be evaluated);

[0149] S1. Obtain the attributes of the reference data sets A1, B1, C1;

[0150] A2’ presents the target data sets a1, b1, c1 and the reference data sets A1, B1, C1 in association;

[0151] Wherein, the target data sets a1, b1, c1 and the reference data sets A1, B1, C1 have correspondence in the (technical) field and / or category to which they belong.

[0152] Further optionally, the target data sets a1, b1, c1 and the reference data sets A1, B1, C1 have the same (technical) field and / or (technical) category.

[0153] Optionally, in some embodiments, the reference data sets A1, B1, C1 have a larger number of elements relative to the target data sets a1, b1, c1.

[0154] Optionally, in some embodiments, the reference data sets A1, B1, C1 have stronger statistical characteristics relative to the target data sets a1, b1, c1 in the corresponding field / category; and / or,

[0155] Optionally, in some embodiments, in the corresponding domain / category, the reference data sets A1, B1, and C1 are more representative than the target data sets a1, b1, and c1.

[0156] Optionally, the data referencing method in some embodiments further includes step A3: clustering / classifying the industry-level data set X1 and the enterprise-level data set Y1 respectively using the same or substantially the same (classification) criteria to obtain multiple reference data sets A1, A1', B1, C1 and multiple target data sets a1, b1, c1. The industry-level data set X1 can be implemented as patent data associated with / related to an industry (e.g., the lithium battery industry), specifically obtained through patent navigation analysis or other methods targeting the relevant industry.

[0157] In some embodiments, reference data sets A1, A1', B1, and C1 can correspond to technical modules and sub-modules (of complete products) decomposed from industry-level data sets, such as positive electrode materials and negative electrode materials in the lithium battery industry. Target data sets a1, b1, and c1, on the other hand, correspond to technical modules and sub-modules (of complete products) decomposed from enterprise-level data sets, such as positive electrode and negative electrode modules involved in the research and development of lithium batteries within CATL. Of course, in other embodiments, the reference data sets A1, A1', B1, and C1 correspond to multiple different complete products decomposed from the industry-level data set, such as BYD's lithium iron phosphate (blade) battery and CATL's ternary lithium battery within the entire lithium battery industry. The target data sets a1, b1, and c1 correspond to the (patent) data of one or more complete products decomposed from the enterprise-level data set. For example, the (patent) data corresponding to the complete battery product of ternary lithium battery involved in the research and development of lithium batteries within CATL, or the (patent) data corresponding to the complete battery product of solid-state battery involved in the research and development of lithium batteries within CATL.

[0158] Of course, as an alternative, in other embodiments, the industry-level data set X1 and the product-level data set can also be clustered / classified respectively in step A3, with the same or substantially the same (classification) criteria, to obtain the multiple reference data sets A1, A1', B1, C1 and the multiple target data sets a1, b1, c1. The product-level data set is a data set for a certain type of product, which can be a cross-company data set, i.e., a data set (e.g., a partial or complete patent or patent application set) generated in the development of one or more different companies producing the same type of product / the same product, in the development of this type of product / this product. Applying the data reference method of the embodiments of the present application to such a product-level data set is equivalent to evaluating the product-level data set with the industry-level data set as the reference / background, so as to reflect whether the product is on the mainstream technical route, as well as the potential technical value, market value, and legal value of the related patent literature, with the help of the industry-level data background.

[0159] The first part data sets A1, B1, C1 generated by clustering / classifying the industry-level data set X1 directly correspond to the data sets a1, b1, c1 generated by clustering / classifying the enterprise-level data set. In order to make the description more concise, the data sets a1, b1, c1 in some embodiments are also directly referred to as enterprise-level data sets, and the first part data sets A1, B1, C1 are also directly referred to as industry-level data sets. Here, the second part data set A1' between the industry-level data set and the enterprise-level data set reflects the difference in composition between the two, and as the difference between the two, the "second part data set" A1' not only has a certain correlation with the first part data sets A1, B1, C1 in the industry-level data set X1, because the data sets A1, B1, C1 and the data set A1' are both generated from the same industry-level data set X1 according to the same or similar criteria, in a popular sense, both belong to branches and branches of the same technology tree, and must have a certain degree of correlation. Alternatively, the second part data set A1' can be regarded as a certain (correlation) extension of the clustering / classification of the first part data sets A1, B1, C1 within the coverage of the industry technology tree, so as to help identify the clustering / classification associated with the enterprise-level data sets A1, B1, C1 and the data therein. Conversely, the industry-level data set X1 including these data sets A1' can in turn serve as a more macro and ordered reference background to provide richer reference dimensions for more accurate evaluation of individual enterprise data sets.

[0160] Of course, it can be understood by those skilled in the art that in some embodiments, the industry-level data set X1 can include an enterprise-level data set, for example, the (patent) data set of the first enterprise. This is because the first enterprise also participates in the research and development activities of the relevant industry and contributes to the formulation of the technology roadmap of the industry and the formation of the technology trend. Of course, the industry-level data set X1 can also be configured to include only the data set of one or more enterprises within the relevant industry, such as the second enterprise and the third enterprise, without including the data set of the first enterprise, that is, there is no intersection between the industry-level data set X1 and the enterprise-level data set, which can increase the contrast effect between the data set of the first enterprise and the data set / union of other multiple enterprises.

[0161] In general, compared with the (single) enterprise-level data set, the industry-level data set X1 includes a larger number of data and more classified data, and can also be divided into more data sub-sets by clustering and the like. For example, taking the lithium battery industry as an example, the industry-level (patent) data set X1 includes about 1 million patent data from all or most of the enterprises in the lithium battery industry, which can be clustered / classified into positive material data set A1, negative material data set A1', lithium recovery technology B1, and electrolyte data set C1. Of course, it can be understood that the data included in the positive material data set A1 is not the patent data of a certain enterprise, but preferably the patent data and patent application data of all or a substantial part of the enterprises in the entire lithium battery industry. Therefore, with reference to the data sets A1, B1, and C1, compared with the positive material data set a1, the lithium recovery technology data set b1, and the electrolyte preparation technology data set c1 of the Ningde Times Company, it is obvious that there is a larger data amount, stronger statistical characteristics, and stronger category representation.

[0162] The enterprise-level (patent) data set of the Ningde Times Company includes about 10,000 data, which can be clustered / classified into positive material data set a1, lithium recovery technology data set b1, and electrolyte preparation technology data set c1. Among them, it is assumed that the research and development work of the negative material of the Ningde Times Company is relatively weak, and it does not include or only includes a small amount of negative material related patents in its enterprise-level patent data set, so the negative material data set cannot be clustered from the enterprise-level data set of the Ningde Times Company, and therefore in FIG. 2, there is no data set in the second list β2 that can correspond to the negative material data set A1' in the first list γ1.

[0163] In addition, it can also be understood that the positive electrode material data set A1 and the negative electrode material data set A1' have a certain correlation relationship, and both belong to a more subdivided subcategory under the large category of electrode materials. They are different from each other, but have a certain relationship in classification / clustering. Therefore, in the first list γ1, preferably, the positive electrode material data set A1 and the negative electrode material data set A1' are arranged adjacently / adjointly, so as to facilitate the user to browse and compare all the technologies in these close positions in a single position in the list C1, C2.

[0164] Optionally, in some embodiments, the enterprise-level data set Y1 is a subset of the industry-level data set X1. In this case, the same standard is used for classification / clustering of both, and the target data sets a1, b1, c1 generated may be subsets of the reference data sets A1, B1, C1. For example, the target data set b1 is a subset of the reference data set B1; the target data set c1 is a subset of the reference data set C1.

[0165] Each enterprise-level data set Y1 can generate multiple target data sets a1, b1, c1 through clustering / classification. The industry-level data set X1 can generate multiple reference data sets A1, B1, C1 through clustering / classification. If the clustering / classification standards are (basically) the same, the reference data set A1 and the target data set a1 have a corresponding relationship in the category, or in other words, the data set A1 and the data set a1 belong to the same category / classification. Similarly, the reference data set B1 and the target data set b1, and the reference data set C1 and the target data set c1 also have corresponding relationships in the category, respectively.

[0166] Note that "(basically) the same" means "the same" or "basically the same". Other bracketed notes have similar explanations and will not be repeated.

[0167] Optionally, in some embodiments, the enterprise-level data set Y1 is a proper subset of the industry-level data set X1.

[0168] Preferably, the industry-level data set X1 comprises data of a plurality of enterprises, while the enterprise-level data set Y1 comprises data of the first enterprise, i.e. the first enterprise's (patent) data is benchmarked against the union / combination of the (patent) data of two or more enterprises. Of course, all of the first enterprise's patent data belongs to the enterprise-level data set Y1, or only part of the first enterprise's patent data belongs to the enterprise-level data set Y1. Similarly, the (partial) inclusion of the first enterprise's patent data in the data set also applies to the "plurality of enterprises" and the industry-level data set X1. Of course, the union / combination of the data of the "two or more enterprises" can or can not include the data of the first enterprise.

[0169] Further, as an alternative, the enterprise-level data set Y1 can comprise the (patent) data of the first enterprise and the second enterprise, while the industry-level data set can comprise the (patent) data of the third enterprise, the fourth enterprise, the fifth enterprise, and so on. Since the industry-level data set comprises the patent data of more enterprises and the trend of market demand of more companies and the (independent) judgment of the evolution of the technical route embodied therein, the industry-level data set X1 and the reference data sets A1, B1, C1 decomposed therefrom form the aggregation of the data of more enterprises and the information or intelligence implied therein, and have stronger statistical characteristics or representativeness in terms of technical trends, market value, etc. The benchmarking of the enterprise-level data set Y1 and the target data sets a1, b1, c1 decomposed therefrom against the industry-level data set and the reference data sets A1, B1, C1 decomposed therefrom to evaluate the patent data of the first enterprise and the second enterprise has reference significance.

[0170] In addition, the plurality of enterprises described above all belong to the industry related to the enterprise-level data set Y1, such as the lithium battery industry in some other embodiments. Further preferably, the plurality of enterprises that can reflect certain trends of the lithium battery industry include at least one leading enterprise in the lithium battery industry, with the "leading" attribute reflected in, for example, the size of the enterprise or the amount of (patent application) data, ranking in the top 10 of the lithium battery industry. Such leading enterprises have industry influence and thus the ability to dominate the technology roadmap, which enables the industry-level data set X1 to better represent the evolution trend and direction of the technology roadmap of the lithium battery industry. Therefore, the reference of the enterprise-level data set Y1 to the industry-level data set X1 has good benchmarking significance and guiding significance for research and development. Of course, even if the plurality of enterprises are not all leading enterprises, but include a part of "non-leading" enterprises, the (patent) data of the plurality of enterprises has strong statistical characteristics or representativeness in the corresponding field / category. Because some small and medium-sized enterprises that are not at the head of the industry often engage in some disruptive and breakthrough research and development directions, to some extent, this can be the starting point and opportunity for new technology roadmaps. Therefore, the reference of the industry-level data set X1 including the (patent) data of such small and medium-sized enterprises to the decomposition, benchmarking and evaluation of the enterprise-level data set Y1 also has benchmarking significance.

[0171] Here, the number of data included in each reference data set A1, B1, C1 is more than the corresponding target data set a1, b1, c1 (to be evaluated). And each reference data set A1, B1, C1 and the target data set a1, b1, c1 have correspondence in classification / clustering, that is, the target data set a1 and the reference data set A1 belong to the same category / clustering, and the number of data samples in the reference data set A1 is (much) more than the number of data samples in the target data set a1. Therefore, the analysis of the reference data set A1, B1, C1 and the indicators generated by the analysis have statistical significance. The target data set divided / clustering from the enterprise-level data set is small in data quantity and lacks statistical significance, so some industry-level trend information cannot be identified by statistical analysis of the small amount of data in the enterprise-level data set Y1, the target data set a1, b1, c1.

[0172] From another perspective, the target data set a1 is a small set, which can belong to the large set of "reference data set" A1, which are both patent technology data of the same category. Based on the small set, it is difficult to analyze statistical information, and based on the large set, it is easier to analyze trend information with statistical significance. Because: "trend" is actually the choice of many companies, or in other words, the large set A1, X1 represents the choice of the majority of companies on the technology route, and then the small set a1, Y1 represents the choice of individual companies or the minority of companies on the technology route. Therefore, through the method of some embodiments of the application, it is realized to take the choice of the majority(major) of companies in the industry on the research and development technology and the research and development data generated thereby as the background to see / evaluate whether the technology route of the minority(minor) of companies conforms to the technology trend of the industry.

[0173] Optionally, in some embodiments, the enterprise-level data set is the patent data of the first enterprise (e.g. Ningde Times) (if the first enterprise is a single-business enterprise). If the first enterprise is a multi-business enterprise, the enterprise-level data set is the patent data of the first enterprise (e.g. BYD) in the first industry (e.g. lithium battery industry), and does not include the patent data related to new energy vehicle whole vehicle owned by BYD company.

[0174] The first enterprise belongs to the first industry, for example, the lithium battery industry. And the industry-level data set is: all or most of the patent data in the first category (e.g. cathode material category) in the lithium battery industry.

[0175] Optionally, in some embodiments, the reference data set A1 includes the patent data of the top N enterprises (e.g. Ningde Times, China Aviation Lithium, LG Chemical, etc.) in the first category (e.g. cathode material category) according to the total number of patents; the top N enterprises belong to the first industry. Then this large set A1 and the small set a1 are provided to the user in association, so that the large set A1 is taken as a reference background for the user to perform benchmarking analysis on the enterprise-level target data set (e.g. the patent data set a1 of the cathode material category of Ningde Times).

[0176] And further optionally, if the industry-level data set X1 is clustered and processed by technology decomposition, a plurality of reference data sets A1, B1, C1 can be generated, wherein the first reference data set A1 can include the patent data of the top N enterprises in the cathode material category according to the total number of patents.

[0177] Of course, in an alternative, the first category can also be the first product category, whereby the patent data of the product level of various products on the market are benchmarked. Of course, it should be understood that in the same industry, there can be multiple products on the market, and in the product level benchmarking process here, the patent data set related to a certain product (such as P1 model or Q1 standard security equipment) can be taken as the target data set (to be evaluated), and the patent data set related to all similar security equipment products including P1 model or Q1 standard in the entire industry can be taken as the reference background to evaluate and assess the technical elements of P1 model or Q1 standard security equipment.

[0178] Optionally, although in the method of some embodiments, the reference data set and the target data set are presented to the user in a correlated manner in step A2, for example, so that the user can more easily correspond or associate them for benchmarking analysis, and it is easier to identify the differences and similarities between the industry-level data set X1 and the enterprise-level data set Y1 as a whole, the reference data set and the target data set are both data sets belonging to different categories such as industry level and enterprise level, and the user also needs to visually distinguish them, so the method of some embodiments can further include the step:

[0179] A8. Independently and / or separately present the reference data set X1 and the target data set Y1 visually, so that the user can more clearly distinguish the reference data set X1 and the target data set Y1 visually.

[0180] Optionally, step A8 further includes the sub-step of presenting the plurality of reference data sets A1, B1', B1, C1 and the plurality of target data sets a1, b1, c1 in the first list γ1 and the second list β2, respectively; the number of sets (e.g. 4) in the plurality of reference data sets A1, B1', B1, C1 is more than the number of sets (e.g. 3) in the plurality of target data sets a1, b1, c1. Optionally, the method of some embodiments further includes step A7: arranging the corresponding target data set a1 and reference data set A1, target data set b1 and reference data set B1, target data set c1 and reference data set C1 in the same row, respectively.

[0181] Specifically, at least part of the target data sets a1, b1 in the second list β2 correspond one-to-one to at least part of the reference data sets A1, B1 in the first list; the corresponding target data set a1 and reference data set A1 are located in the same row LA1. The corresponding target data set b1 and reference data set B1 are located in the same row LB3.

[0182] For example, as shown in FIG. 2, the plurality of target data sets a1, b1, c1 has 3 clusters / classifications, and the plurality of reference data sets A1, B1', B1, C1 has more classifications / clusters, which is 4 in number.

[0183] Optionally, in some embodiments, the plurality of reference data sets A1, B1', B1, C1 has more classifications / clusters than the plurality of target data sets a1, b1, c1; further optionally, the number of sets (e.g. 4) of the plurality of reference data sets A1, B1', B1, C1 is more than the number of sets (e.g. 3) of the plurality of target data sets a1, b1, c1.

[0184] Specifically, the plurality of reference data sets A1, B1', B1, C1 has more data sets than the plurality of target data sets a1, b1, c1. Specifically, the plurality of reference data sets A1, B1', B1, C1 includes two parts: a first part data set A1, B1, C1 and a second part data set B1'. Among them, the data set A1 in the first part data set, the data set B1, and the data set C1 correspond to the data set a1, the data set b1, and the data set c1 in the plurality of target data sets, respectively, in terms of classification / clustering. While the second part data set B1' is a data set (or more data sets) that the plurality of reference data sets has more than the plurality of target data sets.

[0185] Since the first part data set A1, B1, C1 and the second part data set B1' are both data sets obtained by clustering / classifying the industry-level data set X1 in the same standard through step A4 and the like, a) the classification / clustering of the second part data set B1' is different from b) the classification / clustering of the first part data set A1, B1, C1, for example, corresponding to different technical modules, different technical routes. However, at the same time, the first part data set A1, B1, C1 and the second part data set B1' are both differentiated from the (same) industry-level data set X1, and they are related to each other to some extent in terms of classification / clustering, for example, these different technical modules and different technical routes correspond to the same industry, or even the same type of product.

[0186] Further, each data set in the first part data sets A1, B1, C1 can and only can find a unique data set in the target data sets a1, b1, c1 which has a corresponding relationship in classification / clustering, and vice versa. Therefore, this part of the data sets (i.e. the second part data set B1') are different from each other but associated with each other in classification / clustering compared with the target data sets a1, b1, c1.

[0187] In summary, the industry-level data sets X1 are clustered / classified by, for example, step A4, and these clustered / classified data sets A1, A1', B1, C1 are visually structured and presented by step A8 or its sub-steps, so as to visually embody the (hidden) structure in the industry's "technology tree" or the data involved therein in the form of a list. Of course, as an alternative embodiment, the data set A1 and the data set A1' with higher correlation can not be presented nearby, but the second part data set A1' of the clustered / classified data sets A1, A1', B1, C1 which is more than the enterprise-level data sets a1, b1, c1 can be concentrated and separately attached at the tail of the first list γ1 instead of being arranged nearby the data set A1 which is "similar / corresponding in classification / clustering" in the list. Even more, the second part data set A1' is not presented by the first list γ1, but is independently presented to the user by a display area somewhere on the user interface which is irrelevant to the first list γ1.

[0188] Or, in another embodiment, the data sets A1, A1', B1, C1 "generated by industry-level data set X1 clustering / classification" and the data sets a1, b1, c1 "generated by enterprise-level data set Y1 clustering" can also be presented in other visual presentation manners. For example, these data sets A1, A1', B1, C1 are distributed to form a patent map, or a patent landscape, and different colors are used to distinguish the differences in the trend, market value, and other indicators of each data set; or these data sets A1, A1', B1, C1 can also be constructed to form a "technology tree" form, and on the technology tree, the enterprise-level data set a1, the enterprise-level data set b1, and the enterprise-level data set c1 are respectively distributed next to the industry-level data set A1, the industry-level data set B1, and the industry-level data set C1. Therefore, on this patent landscape, it is easier to arrange the i) industry-level patent data sets A1, A1', B1, C1 and the ii) corresponding enterprise-level patent data sets a1, b1, c1 in a more reasonable and more visually understandable contiguous arrangement.

[0189] Of course, it can be understood that in FIG. 2, the third list a3 and the second list b2 are presented side by side, and each corresponding list item in the list is presented in a corresponding row LA1, LC4, LB3. While the data of the first enterprise and the second enterprise are referenced to the industry data and its attributes in the first list g1 as a reference background, the benchmarking between the two will have a stronger industry benchmark, and the benchmarking conclusion will be more meaningful. Further optionally, as the user operates the interface shown in FIG. 2 by mouse or touch screen gesture, the first list g1, the second list b2, and the third list a3 will be moved down synchronously to present the list items in the subsequent rows. The corresponding list items in each list remain aligned at all times, facilitating the user's benchmarking browsing. That is, the items (list items) in the lists g1, b2, and a3 are not limited to being presented at the same time, but can also be presented in groups on the display of the electronic device, and the subsequent list items in the lists g1, b2, and a3 are presented in groups by the user's switching operation. Each group can include one or more data sets in the industry-level data sets A1, A1', B1, C1, one or more data sets in the enterprise-level data sets a1, b1, c1. Or, each group can include one or more data sets in the target data sets a1, b1, c1 in other embodiments, and one or more data sets corresponding thereto in the reference data sets A1, A1', B1, C1. Of course, since these data sets can each include a large amount of data, during the benchmarking display on the display, only the cluster name, category name (such as lithium battery positive material, etc.), or the evaluation index of these sets, etc. Key information can be displayed, as shown in FIG. 2.

[0190] The skilled in the art should understand that the data set A1 and the data set A1' are different but have a certain correlation, and the data set A1 and the data set a1 have the same classification / clustering, so the data set A1' and the data set a1 have a certain correlation. As the above-mentioned examples in some embodiments: the data set A1 is an industry-level patent data set of the lithium battery positive material category, the data set a1 is an enterprise-level (such as Ningde era) patent data set of the lithium battery positive material category, and the data set A1' is an industry-level patent data set of the lithium battery negative material category. The data set A1' and the data set a1 belong to a more subdivided subcategory under the large category of electrode materials of lithium batteries. Ningde era, an enterprise engaged in lithium batteries, may not engage in the research and development of negative materials, or although it engages in relevant research and development work, it ignores the layout of lithium battery negative material related patents. In this case, the user is meaningful to analyze the patent layout of Ningde era company by providing the data set A1' in association with the data set A1 and the data set a1, and Ningde era company may also be interested in the data set A1' related information, because it may help him find the shortcomings in patent layout and research and development activities. This reference presentation method of multiple reference data sets A1, A1', B1, C1 as background data can make the user easily find more patent data subsets A1' that have a certain correlation with the data set of the enterprise in the industry, but the enterprise has not involved in, or even unknown by the enterprise. These may be the research and development directions that the user ignores in daily research and development activities, which may be a valuable technical blank for the enterprise itself, and are also a technical direction worthy of the user's attention and research.

[0191] The data clustering / classification A1' that is different between the industry-level data set X1 and the enterprise-level data set Y1 can reflect the difference between the industry-level data set X1 and the enterprise-level data set Y1, and the patent data set A1' that has a certain correlation with other data in the data sets X1 and Y1. Based on this data set A1', further extension in classification / clustering can help to associate more marginal data of the industry, such as some interdisciplinary data sets. These interdisciplinary data sets also have a certain correlation with the enterprise-level data set Y1, have more macro connections, and are more not limited to direct competitors of the enterprise, so these data can in turn be used as a more macro reference background to evaluate the enterprise data and provide more extensive research and development inspiration and guidance to individual users of the enterprise.

[0192] From another aspect, the data sets in the first part data sets A1, B1, C1 correspond to the data sets in the multiple target data sets a1, b1, c1 in classification / clustering, respectively.

[0193] Optionally, the method of some embodiments further comprises the following steps A5, A6:

[0194] Step A5, arranging the plurality of reference data sets A1, A1', B1, C1 in the first list γ1 in proximity according to the similarity of classification / clustering. Optionally, in step A5, the industry-level data sets can be clustered from the perspective of the functional category, product category, or application category of the technical solutions disclosed in the patent data, and the plurality of reference data sets generated by clustering can be aggregated / classified and presented, and similar categories are arranged in adjacent display positions. In step A6, the data set A1 and the data set A1' that are closer in classification / clustering are arranged closer in the first list γ1.

[0195] This means that the data sets A1, A1' that are more similar and closer in classification / clustering are arranged closer in the first list γ1. The measurement of the similarity, similarity, and closeness between the data sets A1, A1' can be obtained by applying the Euclidean distance, cosine similarity, and other such calculation methods commonly used in the industry to these data sets.

[0196] Optionally, in the method of some embodiments, the reference data sets A1, A1', B1, C1 are reference patent data sets, and the target data sets a1, b1, c1 are target patent data sets.

[0197] Optionally, in the method of some embodiments, the industry-level data sets A1, A1', B1, C1 are industry-level patent data sets, and the enterprise-level data sets a1, b1, c1 are enterprise-level patent data sets.

[0198] Optionally, in some embodiments, the attributes of the enterprise-level data set Y1, the industry-level data set X1, or the reference patent data sets A1, A1', B1, C1 include at least one of the following: market value degree, technical value degree, legal value degree, and cross-disciplinary degree.

[0199] Optionally, in the method of some embodiments, one of the attributes of the target patent data sets a1, b1, c1 or the reference patent data sets A1, A1', B1, C1 is a trend feature; the trend feature is a market trend degree, and at least partially represents / refiects the market value of the technical category corresponding to the reference patent data sets A1, A1', B1, C1.

[0200] The trend feature is a technical trend degree, and at least partially represents the technical value of the target patent data sets a1, b1, c1 or the reference patent data sets A1, A1', B1, C1.

[0201] Optionally, the method of some embodiments or step A2 therein, further comprises the following step:

[0202] The plurality of reference data sets A1, A1', B1, C1 are arranged in the first list in descending order / ascending order of the evaluation indexes of the reference data sets A1, A1', B1, C1. This facilitates efficient browsing according to the weight of the indexes.

[0203] Statistical analysis is performed on the plurality of reference data sets A1, A1', B1, C1 to obtain one or more evaluation indexes, such as "medium technical activity", "high technical advancement", "high market value", etc. as shown in the first list γ1 of FIG. 2.

[0204] Generally speaking, technical trends are driven by large companies (e.g. the top N companies in the industry), which also define the relevant technical routes in the (patent) legal space through a large number of patent applications related to the technical solutions. Therefore, the statistical analysis of the patent data of large companies, especially those with a large number of patent applications, can largely reflect the technical trends and the extension direction of the technical routes in the industry. In addition, trends can be reflected in the technical routes selected by a large number of companies in the industry, the R&D direction selected by the majority of companies, etc. Therefore, the analysis of the increment and stock of patent data of a certain enterprise or industry can also largely reflect the trends. In particular, by comparing the data sets A1, A1', B1, C1 of different clusters in the industry, it can be identified which technical routes are more likely to be chosen by more companies and more likely to be successful, because different reference data sets A1, A1', B1, C1 may represent different technical routes corresponding to the patent data.

[0205] Optionally, the method of some embodiments further comprises the step of: based on the indexes of the reference patent data sets A1, A1', B1, C1, assigning an evaluation index corresponding to the similarity / similarity degree between the target patent data set a1, b1, c1 and the reference patent data set to the target set a1, b1, c1.

[0206] Optionally, in the method of some embodiments, the similarity / similarity degree can be measured by the proportion of the intersection of i) the target patent data set a1 and ii) the reference patent data set A1 in the target patent data set a1; or,

[0207] iii) the proportion of the intersection of iii) the target patent data set b1 and iv) the reference patent data set B1 in the reference patent data set B1; or,

[0208] a) the size of the intersection of the target patent data set c1 and the reference patent data set C1, and b) the size of the union of the target patent data set c1 and the reference patent data set C1.

[0209] For example, the relative proportion relationship between a) one reference data set A1 and b) one target data set a1 in FIG. 1b is described, wherein the reference data set A1 is one of the plurality of reference data sets A1, A1', B1, C1 generated by clustering the industry-level data sets, and the target data set a1 is one of the plurality of target data sets a1, b1, c1 generated by clustering the enterprise-level data sets. Here, it is assumed that the number of data elements in the reference data set A1 in FIG. 1b is 100, and the number of data elements in the target data set a1 is 80, and the target data set a1 is a proper subset of the reference data set A1, so the correlation between them is strong, and the above-mentioned proportion and ratio also exceed the first predetermined threshold. Here, if the evaluation index obtained by statistical analysis of the reference data set A1 is "high market value" and "high interdisciplinary degree", the target data set a1 can also be assigned the evaluation index of "high market value" and "high interdisciplinary degree".

[0210] Of course, the similarity, the similarity, and the distance involved in the "proximity arrangement" of the data set in some embodiments can also be measured by means of Euclidean space distance, and if the Euclidean distance is less than a certain threshold, it is considered that the two data sets are close.

[0211] The method of some embodiments of the present application can be applied to patent databases for patent analysis, and in related technical fields and markets, incopat (Xinzhao Company), patsnap (Zhihui Me Company), patentics (Soyi Interactive) and other enterprises can use this method to improve the analysis effect / efficiency, interactive effect of their patent analysis, and provide better performance (Feature) for their products. The semantic analysis function of the patentics (Soyi Interactive) software is very powerful, and with the help of its semantic analysis capability and the text processing capability of Deepseek, the method of the embodiments of the present application can achieve better data reference effect.

[0212]

Alternative Embodiment

[0213] Optionally, in some embodiments of the present application, step A8 further comprises:

[0214] The plurality of reference data sets and the plurality of target data sets are respectively presented in a first list and a second list; and at least part of the target data sets in the second list are one-to-one corresponding to at least part of the reference data sets in the first list.

[0215] Preferably, the industry-level data, the enterprise-level data, are patent data of the last M years; wherein, M = 2, 3, 5, 8, or 10. Wherein, A) the industry-level data set is patent data of a single country; and / or, B) the industry-level data are not patent data of all countries / regions in the world, but only include patent data of a single country, or patent data of multiple countries.

[0216] It should be noted that the order of the steps in some embodiments of the present application is not limited. For example, step A0' can occur before or after step S1, that is, in some embodiments, the reference data set can be provided to the target data set first, and then some attributes of the reference data set are obtained; while in some other embodiments, some attributes of the reference data set can be obtained first, and then the reference data set is provided to the target data set.

[0217] The numbers of the steps do not represent the order in time. In addition, more variations due to the change of the order of the steps will not be described in detail.

[0218] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above drawings are used to distinguish similar objects, and do not necessarily describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein.

[0219] Each embodiment in the specification is described in a progressive and complementary manner, and the same or similar parts between each embodiment can be referred to each other. Each optional technical feature can be combined with other embodiments in any reasonable manner. The content between each embodiment and under each title can also be reasonably combined. Each embodiment focuses on the difference from other embodiments. In particular, for system embodiments, since they are basically similar to method embodiments, the description is relatively simple, and the relevant parts can be referred to the part of the method embodiment.

[0220] The terms used in the embodiments of the present application are merely for the purpose of describing specific embodiments and are not intended to limit the present application. The singular forms "a," "said," and "the" used in the embodiments of the present application and the appended claims are also intended to include the plural forms, unless the context clearly indicates otherwise. "Multiple" generally includes at least two. It should be understood that the term "and / or" used herein merely describes an association relationship of associated objects, which means that there can be three relationships, for example, A and / or B can represent the three cases of A alone, A and B together, and B alone. In addition, the character " / " herein generally represents an "or" relationship between the front and rear associated objects.

[0221] Although the specific embodiments of the present application are described above, those skilled in the art should understand that this is only an example, and the protection scope of the present application is defined by the appended claims. Those skilled in the art can make various changes or modifications to these embodiments without departing from the principles and essence of the present application, and such changes and modifications fall within the protection scope of the present application.

[0222] In recent years, with the explosive growth of the network live broadcast industry, live broadcast platforms have become an important carrier for content dissemination, business promotion and social interaction. According to statistics, the user scale of China's live broadcast industry has broken through 700 million in 2023, and the market size has reached the level of 100 billion yuan. However, with the rapid development of the industry, the head platforms generally use technical means to build a "data moat" to consolidate their market position, which is manifested in:

[0223] Information monitoring barrier: Through technologies such as sensitive word filtering and multi-modal content recognition (ASR / OCR), real-time interception of information related to competitor platforms (such as keywords such as "WeChat diversion" and "Kuaishou anchor"), and even detection capabilities for traditional evasion methods such as voice modulation and similar character replacement;

[0224] Protocol-level control: Deep analysis of transmission protocol payloads (such as RTMP data packets), feature scanning of structured data and hidden fields;

[0225] Behavior pattern analysis: Based on user interaction timing, input frequency and other features, machine learning models are established to identify irregular communication behavior.

[0226] Such technical monopolistic behavior leads to two major contradictions:

[0227] User rights are damaged: anchors cannot reasonably divert across platforms (such as mentioning WeChat customer service in e-commerce live broadcasts), and businesses are forced to accept high platform commission;

[0228] Industry ecological imbalance: Small and medium-sized platforms are difficult to participate in fair competition due to traffic blockade, forming a "winner-takes-all" market pattern.

[0229] Existing circumvention solutions have significant drawbacks:

[0230] Static replacement rules: fixed vocabulary is easily covered by platform feature libraries and cannot adapt to dynamically updated detection strategies;

[0231] Single-modal processing limitations: only supports text replacement, lacks the ability to respond to platform upgrades such as voice fingerprint recognition and video frame analysis;

[0232] Vulnerable synchronization mechanism: relies on centralized server synchronization of encryption and decryption rules, with single point of failure and traceability risks;

[0233] Feature exposure risks: artificially designed replacement patterns are easily captured by machine learning models (such as fixed position insertion of interference symbols).

[0234] In addition, competition barriers between platforms also promote technological monopolistic behavior. For example, Douyin indirectly restricts users' discussions of other platforms by prohibiting the mention of competitor platform names, while platforms such as Kuaishou further tighten content review through the "no broadcast without record" policy. While this regulatory model helps maintain the platform ecosystem, it may also inhibit innovation and user diversity.

[0235] In summary, the existing regulatory technology of live streaming platforms has problems such as poor dynamic adaptability, insufficient multi-modal processing capability, and lack of context understanding, and there is an urgent need for a technical solution that can both circumvent sensitive word monitoring and protect user freedom of expression. Therefore, developing an information processing method based on dynamic replacement rules, context awareness, and end-to-end encryption synchronization has become a key requirement to solve the above technical problems.

[0236] Therefore, in another embodiment, a method for processing information (such as voice input) in a live stream is also provided, including the following steps:

[0237] SS1). (At the first user terminal) collect the input information (such as voice, text, etc.) of the first user; SS2). Replace a specific part of the input information with a non-sensitive word (which can be set arbitrarily, such as Apple, Taishan, etc.) of the current live streaming platform; SS3). Transmit the input information (after partial replacement) to the current live streaming platform.

[0238] Optionally, the information processing method further comprises: SS4) receiving, at the second user terminal, the input information from the current live platform, replacing the non-sensitive words in the input information back to the specific part, thereby restoring / recovering the input information as the original input content of the first user, and outputting the (original input information with the specific part) to the second user. And / or, preferably, the rules for replacing / substituting the specific part of information are synchronized between the first user terminal and the second user terminal, so that both use the same rules for replacement / substitution (first user terminal) and recovery / restoration (second user terminal). Alternatively, the first user terminal and the second user terminal use the same codebook, and both replace and restore the specific part according to the same codebook, thereby avoiding monitoring and checking of the specific information in the current platform during the transmission of the input information via the current platform. In this way, by replacing the platform-sensitive specific part in the input information at the first user terminal, the input information will not be monitored and warned by the platform during the transmission via the live platform, thus avoiding some exclusive and monopolistic behaviors / policies of the current platform (for example, the Douyin platform does not allow the live streamer as the first user to mention the name of other platforms during live streaming).

[0239] Optionally, the specific part is a sensitive word of the current live platform (such as Douyin), for example, these sensitive words represent the names of other second video platforms or second live platforms, such as WeChat, Kuaishou, (Sina) Weibo, etc.

[0240] The following is a further optimization and refinement of the scheme for avoiding monitoring of sensitive words such as the names of other platforms by the live platform, from the dimensions of technical implementation, security, scalability, legal compliance, etc., to enhance the innovativeness and practicality of the scheme:

[0241] I. Optimization goal

[0242] 1. Enhance concealment: avoid detection of replacement rules by the platform.

[0243] 2. Improve flexibility: support dynamic adjustment of replacement strategies to adapt to different platform sensitive word policies.

[0244] 3. Ensure security: prevent reverse engineering or cracking of replacement rules.

[0245] 4. Expand application scenarios: applicable to voice, text, image, etc. multi-modal input.

[0246] 5. Legal compliance: ensure that the technical means do not violate platform rules or relevant laws and regulations.

[0247] II. Optimization and refinement scheme

[0248] 1. Dynamic replacement rules and context awareness

[0249] (1) Context-based intelligent replacement

[0250] - Problem: Fixed replacement words (such as "apple" and "Tayshan") are easy to be discovered by platform anti-crawler or AI detection.

[0251] - Optimization:

[0252] - Deploy context-aware algorithms on the first user terminal to dynamically select replacement words based on the semantics of input information.

[0253] - For example:

[0254] - If the input is "go to WeChat to chat", the system can analyze the context intention (such as "social") and dynamically replace it with "go to Oasis to chat" ("Oasis" is a social product in the Douyin ecosystem).

[0255] - If the input is "fast live is very popular", the system can replace it with "extreme live is very popular" combined with current popular vocabulary ("extreme" is a function under Douyin).

[0256] (2) Multi-layer replacement strategy

[0257] - Problem: Single replacement rule is easy to be cracked.

[0258] - Optimization:

[0259] - Introduce multi-level replacement strategy, for example:

[0260] 1. Primary replacement: replace sensitive words with platform-allowed synonyms (such as "platform" instead of "Douyin").

[0261] 2. Secondary replacement: further replace synonyms with fuzzification words (such as "platform" instead of "this application").

[0262] 3. Encryption replacement: encode fuzzification words into symbols or pinyin initials (such as "this application" instead of "zgyy").

[0263] 2. Encryption and synchronization mechanism

[0264] (1) End-to-end encrypted replacement rules

[0265] - Problem: If the replacement rules are stored or transmitted in plaintext, they are easy to be analyzed inversely by the platform.

[0266] - Optimization:

[0267] - Establish an encrypted channel between the first user terminal and the second user terminal to synchronize the replacement rules.

[0268] - Replace rules are stored in JSON format encrypted with a key, only decrypted at the terminal.

[0269] (2) Timestamp synchronization

[0270] - Problem: If the replacement rules are fixed for a long time, they are easy to be captured by the platform.

[0271] - Optimization:

[0272] - Update the replacement rules every fixed time (e.g. 1 hour), and ensure the same rules are used on both ends through timestamp synchronization mechanism. Preferably, update the sensitive word library and replacement rules regularly.

[0273] - Example:

[0274] - The first terminal generates a rule version number `V_20250507_12`, and the second terminal automatically matches the rules according to the timestamp.

[0275] 3. Multi-modal input processing

[0276] (1) Joint processing of speech and text

[0277] - Problem: Existing solutions only target text and cannot handle speech input.

[0278] - Optimization:

[0279] - Add an ASR module in SS1 to convert speech input to text before replacement.

[0280] - Add a TTS module in SS4 to convert the restored text to speech output.

[0281] - Example flow:

[0282] Speech input → ASR → Text → Replacement → Transmission → Restoration → TTS → Speech output

[0283] (2) Steganography in images and videos

[0284] - Problem: Live may involve image or video content (such as screen sharing competitor platform interface).

[0285] - Optimization:

[0286] - Embed steganography in images / videos, encode sensitive information as pixel perturbation or watermark.

[0287] - For example: Replace the "WeChat" icon with the "TikTok" icon, but hide the original information by fine-tuning the pixel values.

[0288] 4. Legal compliance design

[0289] (1) Avoid platform rules without violating the law

[0290] - Risk point: Directly bypassing platform monitoring may be considered as a violation.

[0291] - Optimization:

[0292] - Clearly not generate false information when designing, only through linguistic skills to avoid sensitive word detection.

[0293] - For example: Replace "Kuaishou" with "another short video platform" instead of directly mentioning the competitor's name.

[0294] (2) User informed consent mechanism

[0295] - Risk point: Users may not know that information has been replaced.

[0296] - Optimization:

[0297] - Display transparent prompt on the first user terminal, informing users that "part of the information has been optimized to comply with platform specifications".

[0298] - Provide restoration confirmation on the second user terminal to ensure that users are aware that the information has been restored to the original content.

[0299] 5. Extend application scenarios

[0300] (1) Cross-platform collaboration

[0301] - Problem: Existing solutions only target a single platform (such as Douyin).

[0302] - Optimization:

[0303] - Design a universal replacement rule library that adapts to the sensitive word policies of multiple platforms.

[0304] - For example: Different sensitive word lists for Douyin, Kuaishou and Weibo, dynamically load corresponding replacement strategies.

[0305] (2) Combined with blockchain

[0306] - Problem: Replacement rule synchronization relies on centralized servers.

[0307] - Optimization:

[0308] - Use blockchain to store replacement rule hash values to ensure that rules cannot be tampered with.

[0309] - Example flow:

[0310] - The first terminal uploads the rule hash to the blockchain.

[0311] - The second terminal synchronizes after verifying the legality of the rules through the blockchain.

[0312] Another embodiment provides an information processing method, comprising the following steps:

[0313] 1. Collecting input information of a first user at a first user terminal;

[0314] 2. Replacing specific parts of the input information with non-sensitive words of a current live streaming platform according to preset dynamic replacement rules;

[0315] 3. Transmitting the partially replaced input information to the current live streaming platform;

[0316] 4. Receiving the input information from the current live streaming platform at a second user terminal and replacing the non-sensitive words back to the original content according to synchronized dynamic replacement rules, and outputting to a second user.

[0317] Optionally, the dynamic replacement rules are generated based on context semantic analysis and synchronized between the first user terminal and the second user terminal through an encrypted channel.

[0318] Optionally, the input information includes one or more of voice, text, and image, and is converted between voice and text through voice recognition and synthesis.

[0319] Optionally, the generation of the non-sensitive words includes:

[0320] a. Primary replacement: replacing sensitive words with platform-allowed synonyms;

[0321] b. Secondary replacement: replacing synonyms with fuzzified words;

[0322] c. Encryption replacement: encoding fuzzified words into symbols or pinyin initials.

[0323] Four, technical advantages

[0324] 1. High concealment: dynamic replacement rules and context-aware technology make it difficult for AI to detect the replaced content.

[0325] 2. Strong security: encryption synchronization and blockchain technology ensure that the rules are not tampered with.

[0326] 3. Multi-modal support: covering multiple input forms such as voice, text, and image.

[0327] 4. Legal compliance: avoid legal risks through transparent prompts and user informed consent.

[0328] Five, potential application scenarios

[0329] 1. Live streaming of goods: the host does not trigger a sensitive word warning when mentioning a competitor's platform.

[0330] 2. Content creation: creators avoid content censorship when discussing other platform features.

[0331] 3. Cross-platform collaboration: internal communication across platforms in enterprises avoids information leakage.

[0332] 4. Financial technology scenarios:

[0333] -1) Cross-border investment information sharing

[0334] Scenario: Hong Kong stock analysts circumvent discussion restrictions on exchanges such as "Binance" and "Coinbase"

[0335] Implementation: Encode exchange names into stock code patterns (e.g. "BNB.US" → "02618.HK")

[0336] Risk control mechanism: Establish a dynamic verification system for financial compliance vocabulary.

Claims

1. A method of data referencing, comprising: A0. providing a reference data set, a target data set; S1. obtaining attributes of the reference data set; A2. evaluating corresponding attributes of the target data set, with reference to the attributes; wherein the target data set and the reference data set have correspondence in the field and / or category to which they belong.

2. A method of data referencing, comprising: A0’ providing a reference data set for a target data set; S1. obtaining attributes of the reference data set; A2’ presenting i) the target data set in association with ii) the reference data set and the attributes; wherein the target data set and the reference data set have correspondence in the field and / or category to which they belong.

3. The method of data referencing of claim 1 or 2, wherein, the reference data set has a greater number of elements relative to the target data set; or, has stronger statistical characteristics or representativeness in the corresponding field / category.

4. The method of data referencing of claim 3, wherein, the method further comprises step A4: I) clustering / classifying industry-level data sets and enterprise-level data sets respectively to obtain a plurality of the reference data sets and a plurality of the target data sets, with the same standard / rule / algorithms; or, II) clustering / classifying industry-level data sets and product-level data sets respectively to obtain a plurality of the reference data sets and a plurality of the target data sets, with the same standard / rule / algorithms; or, III) clustering / classifying industry-level data sets and product-level data sets respectively to obtain a plurality of the reference data sets and a plurality of the target data sets; IIII) clustering / classifying industry-level data sets and enterprise-level data sets respectively to obtain a plurality of the reference data sets and a plurality of the target data sets; wherein the plurality of the target data sets correspond to at least part of the plurality of the reference data sets in classification / category respectively; and / or, the plurality of target data sets are different data sets distinguished by functional category / product category / application category; the plurality of reference data sets are different data sets distinguished by functional category / product category / application category.

5. The method of data referencing of claim 4, wherein, the enterprise-level data set is a proper subset of the industry-level data set; and / or, the target data set is a proper subset of the reference data set; and / or, the reference data set comprises data of a plurality of enterprises, and the target data set is data of a first enterprise; the plurality of enterprises and the first enterprise all belong to a first industry; the plurality of enterprises comprises at least one leading enterprise in the first industry; the leading enterprise is among the top 5, top 10, or top 30 in the first industry in terms of market share, market size, or patent application amount.

6. The method of data referencing of claim 5, wherein, the enterprise-level data set is patent data owned by the first enterprise and belonging to the first industry; or, ​ The enterprise-level data set is patent data of the first enterprise in a first category; The first enterprise belongs to a first industry, and the industry-level data set is all or part of patent data of the first industry in the first category or all or part of patent data of the first industry in the first category.

7. The method of data referencing of claim 6, wherein, The industry-level data set includes patent data of the top N enterprises in the first category in terms of total number of patents; the top N enterprises belong to the first industry; or the industry-level data set includes patent data of the top N enterprises in the first industry in terms of total number of patents, and N is 5, 10, 20 or 30; and / or, The first category is a first function category, a first product category, or a first application category.

8. The method of data referencing of any one of claims 1-4, further comprising the step of: A8. presenting the reference data set and the target data set visually independently and / or separately.

9. The method of data referencing of claim 8, wherein, The step A8 further comprises: presenting a plurality of the reference data sets and a plurality of the target data sets in a first list and a second list, respectively; and, corresponding one-to-one at least part of the target data set in the second list to at least part of the reference data set in the first list.

10. The data referencing method as described in claim 9, wherein, The method or the step A2' further comprises a step A7: arranging the target data set and the reference data set corresponding to each other in the same row; wherein the plurality of the reference data sets has more categories / clusters than the plurality of the target data sets; and / or, the number of sets in the plurality of the reference data sets is greater than the number of sets in the plurality of the target data sets; and / or, The plurality of reference data sets includes a first part data set and a second part data set; The categories / clusters of each data set in the first part data set correspond to the categories / clusters of each data set in the plurality of target data sets, and the categories / clusters of each data set in the plurality of target data sets correspond to the categories / clusters of each data set in the first part data set; or, The plurality of the target data sets correspond one-to-one to the first part data set in terms of categories / clusters; or, The plurality of the target data sets correspond one-to-one to the first part data set in terms of categories / clusters, respectively; or, The categories / clusters of each data set in the first part data set correspond to the categories / clusters of one of the plurality of the target data sets, and the categories / clusters of each data set in the plurality of the target data sets correspond to the categories / clusters of one of the first part data sets; or, The plurality of the target data sets correspond one-to-one to the first part data set in terms of categories / clusters; or, The plurality of the target data sets correspond one-to-one to the first part data set in terms of categories / clusters, respectively.

11. The method of claim 10, wherein, i) the classification / cluster of the second part of data sets is different from and is correlated to ii) the classification / cluster of the plurality of the target data sets; or, a) the classification / cluster of the second part of data sets is different from and is correlated to b) the classification / cluster of the first part of data sets.

12. The method of claim 11, further comprising the steps of: arranging the plurality of the reference data sets in the first list in proximity of the classified / clustering similarity; or, arranging the data sets in the first list in a closer manner according to their classification / cluster; or, arranging the data sets in the first list in a closer manner according to their classification / cluster; and / or, the attribute is an evaluation index, arranging the plurality of the reference data sets in the first list according to the descending / ascending order of the evaluation index of the reference data sets; statistically analyzing the plurality of the reference data sets respectively to obtain the evaluation index of each of the plurality of the reference data sets.

13. The method of data referencing of claim 12, wherein, the reference data sets are reference patent data sets and the target data sets are target patent data sets; or, the data elements in the reference data sets are reference patent data and the data elements in the target data sets are target patent data; the evaluation index includes market value, or technical value, or legal value; or, the evaluation index of the reference data sets is a trend characteristic; the trend characteristic includes: cross-disciplinary degree, and / or, market trend degree, and / or, technical trend degree; the market trend degree at least partially represents the market value of the technical category corresponding to the reference data sets; and / or, the technical trend degree at least partially represents / reflects the technical value of the technical category corresponding to the reference data sets.

14. The method of data referencing of claim 13, wherein, further comprising the steps of: assigning a target evaluation index to the target data sets according to the evaluation index of the reference data sets; or, assigning a target evaluation index to the target data sets according to the correlation / similarity between the target data sets and the reference data sets based on the evaluation index of the reference data sets.

15. The method of data referencing of claim 14, wherein, the correlation or similarity is: i) the proportion of the intersection of ii) the target data sets and the reference data sets in the target data sets; or, iii) the proportion of the intersection of iii) the target data sets and the reference data sets in the reference data sets; or, a) the ratio of the size of the intersection of the target data sets and the reference data sets to b) the size of the union of the target data sets and the reference data sets; the first list and the second list are arranged in a longitudinal or transverse manner, side by side / parallel / parallelly; the industry-level data sets and the enterprise-level data sets are patent data of the last M years; wherein, M = 2, 3, 5, 8, or 10; and / or, the industry-level data sets are patent data of a single country.

16. An electronic device, comprising: a display. a processor unit; a memory; and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the processor unit, the one or more programs including instructions for performing any of the methods of claims 1-15.

17. An information processing apparatus for use in an electronic device with one or more processors, comprising: means for performing any of the methods of claims 1-15.

18. The information processing apparatus of claim 17, further comprising means for outputting, via the display, at least one interface or at least one option configured in the electronic device.

19. A non-transitory computer readable storage medium storing one or more programs, the one or more programs comprising instructions, which when executed by one or more processors of an electronic device, cause the electronic device to perform any of the methods of claims 1-15.

20. The non-transitory computer readable storage medium of claim 19, the one or more programs further comprising instructions executable by the electronic device to output, via a display, at least one interface or at least one option configured in the electronic device.

21. An electronic device, comprising: a display, a processor unit and a touch-sensitive surface unit configured to detect contact; wherein the processor unit is coupled with the display and the touch-sensitive surface unit, the processing unit being programmed to perform the steps of any of the methods of claims 1-15.

22. A method of data referencing, comprising: S7. providing enterprise-level data and corresponding industry-level data; S2. structurally presenting the industry-level data; S3. referencing the structured industry-level data, presenting the enterprise-level data.

23. The method of data referencing of claim 22, wherein, Further comprising steps of: S4. obtaining attributes of the industry-level data; S5. referencing the attributes, defining / determining corresponding attributes of the enterprise-level data.

24. The method of data referencing of claim 23, wherein, a) the step S2 further comprises a step S2-1A: clustering / classifying / decomposing / technically decomposing the industry-level data according to a first criterion / first rule / first algorithm to obtain one or more reference data sets; and / or, clustering / classifying / decomposing / technically decomposing the enterprise-level data according to a first criterion / first rule / first algorithm to obtain one or more target data sets; or, b) the step S2 further comprises a step S2-1B: clustering / classifying / decomposing / technically decomposing the industry-level data to obtain one or more reference data sets; and / or, clustering / classifying / decomposing / technically decomposing the enterprise-level data to obtain one or more target data sets; or, c) the step S2 further comprises a step S2-1C: clustering / classifying / decomposing / technically decomposing first product-related data among the industry-level data to obtain one or more reference data sets; and / or, clustering / classifying / decomposing / technically decomposing the enterprise-level data to obtain one or more target data sets; or, ​ ​ clustering / classifying / resolving / technically resolving the first product-related data in the enterprise-level data to obtain one or more target data sets; wherein the clustering is according to the same standard / rule / algorithm; or the classifying is according to the same standard / rule / algorithm; or the resolving / technically resolving is according to the same standard / rule / algorithm; or the multiple target data sets are different data sets distinguished by function category / product category / application category; the multiple reference data sets are different data sets distinguished by function category / product category / application category.

25. The method of data referencing of claim 24, wherein, The step S5 further comprises: determining the attribute of the target data set corresponding in field and / or category according to the attribute of the reference data set; or determining the attribute of the first target data set in the one or more target data sets according to the attribute of the first reference data set in the one or more reference data sets; wherein the first reference data set and the first target data set correspond in field and / or category; and / or The data reference method or the step S3 further comprises a step S6: presenting i) the multiple target data sets and ii) the multiple reference data sets and their attributes corresponding in field and / or category in association.

26. The data reference method of claim 25, wherein the enterprise-level data is patent data owned by the first enterprise and belonging to the first industry; or the enterprise-level data is patent data of the first enterprise in the first category; the first enterprise belongs to the first industry; the industry-level data is i) all or part of patent data of the first industry in the first category; or ii) all or part of patent data belonging to the first industry in the first category. the enterprise-level data is a proper subset of the industry-level data; and / or the target data set is a proper subset of the reference data set of the corresponding category; 27. The method of data referencing of claim 26, wherein, and / or the reference data set comprises data of multiple enterprises, and the target data set is data of the first enterprise; the multiple enterprises and the first enterprise all belong to the first industry; the multiple enterprises comprise at least one leading enterprise in the first industry; the leading enterprise is among the top 5, top 10, or top 30 in the first industry in terms of market share, market size, or number of patent applications. the industry-level data comprises patent data of the top N enterprises in the first category in terms of total number of patents; or 28. The method of data referencing of claim 27, wherein, the first reference data set comprises patent data of the top N enterprises in the first category in terms of total number of patents; the top N enterprises all belong to the first industry; or the industry-level data set comprises patent data of the top N enterprises in the first industry in terms of total number of patents, and N is 5, 10, 20, or 30; and / or the first category is a first function category, a first product category, or a first application category. ​ 29. The method of data referencing according to any one of claims 22-28, wherein the step S3 further comprises a step of: A8. presenting I) the one or more reference data sets, II) the one or more target data sets, visually independently and / or separately.

30. The method of data referencing of claim 29, wherein, The step A8 further comprises: presenting I) the one or more reference data sets, II) the one or more target data sets, in a first column and a second column respectively; the first column is arranged side by side / parallel / parallel with the second column and is presented synchronously; and, presenting the plurality of target data sets in the second column one by one corresponding to part or all of the plurality of reference data sets in the first column.

31. The method of data referencing of claim 30, wherein, The method of data referencing or the step S6 further comprises a step A7: aligning / arranging the mutually corresponding data sets among the plurality of reference data sets and the plurality of target data sets in the same row; wherein the plurality of reference data sets has more categories / clusters than the plurality of target data sets; and / or, the number of sets in the plurality of reference data sets is more than the number of sets in the plurality of target data sets; and / or, the plurality of reference data sets comprises a first part of data sets and a second part of data sets; and the categories / clusters of each data set in the first part of data sets respectively correspond to the categories / clusters of each data set in the plurality of target data sets, and the categories / clusters of each data set in the plurality of target data sets respectively correspond to the categories / clusters of each data set in the first part of data sets; or, the plurality of target data sets one by one correspond to the categories / clusters of each data set in the first part of data sets; or, the plurality of target data sets respectively one by one correspond to the categories / clusters of each data set in the first part of data sets.

32. The method of data referencing according to claim 31, wherein, i) the categories / clusters of the second part of data sets are different from but associated with ii) the categories / clusters of the plurality of target data sets; or, a) the categories / clusters of the second part of data sets are different from but associated with b) the categories / clusters of the first part of data sets.

33. The method of data referencing according to claim 32, further comprising a step of: arranging the plurality of reference data sets in the first column according to the proximity of the categories / clusters; or, arranging the data sets in the plurality of reference data sets that are closer in the categories / clusters closer in the first column; or, arranging the data sets in the plurality of reference data sets that are close in the categories / clusters closer in the first column; and / or, the attribute is an evaluation index, arranging the plurality of reference data sets in the first column according to the descending / ascending order of the evaluation index of the plurality of reference data sets; and / or, the attribute is a classification index, arranging the plurality of reference data sets in the first column according to the descending / ascending order of the classification index of the plurality of reference data sets; and / or, the attribute is a classification index, arranging the plurality of reference data sets in the first column according to the descending / ascending order of the classification index of the plurality of reference data sets; and / or, the attribute is a classification index, arranging the plurality of reference data sets in the first column according to the descending / ascending order of the classification index of the plurality of reference data sets; and / or, the attribute is a classification index, arranging the plurality of reference data sets in the first column according to the descending / ascending order of the classification index of the plurality of reference data sets; and / or, the attribute is a classification index, arranging the plurality of reference data sets in the first column according to the descending / ascending order of the classification index of the plurality of reference data sets; and / or, the attribute is a classification index, arranging the plurality of reference data sets in the first column according to the descending / ascending order of the classification index of the plurality of reference data sets; and / or, the attribute is a classification index, arranging the plurality of reference data sets in the first column according to the descending / ascending order of the classification index of the plurality of reference data sets; and / or, the attribute is a classification index, arranging the plurality of reference data sets in the first column according to the descending / ascending order of the classification index of the plurality of reference data sets; and / or, the attribute is a classification index, arranging the plurality of reference data sets in the first column according to the descending / ascending order of the classification index of the plurality of reference data sets; and / or, the attribute is a classification index, arranging the plurality of reference data sets in the first column according to the descending / ascending order of the classification index of the plurality of reference data sets; and / or, the attribute is a classification index, arranging the plurality of reference data sets in the first column according to the descending / ascending order of the classification index of the plurality of reference data sets; and / or, the attribute is a classification index, arranging the plurality of reference data sets in the first column according to the descending / ascending order of the classification index of the plurality of reference data sets; and / or, the attribute is a classification index, arranging the plurality of reference data sets in the first column according to the descending / ascending order of the classification index of the plurality of reference data sets; and / or, the attribute is a classification index, arranging the plurality of reference data sets in the first column according to the descending / ascending order of the classification index of the plurality of reference data sets; and / or, the attribute is a classification index, arranging the plurality of reference data sets in the first column according to the descending / ascending order of the classification index of the plurality of reference data sets; and / or, the attribute is a classification index, arranging the plurality of reference data sets in the first column according to the descending / ascending order of the classification index of the plurality of reference data sets; and / or, the attribute is a classification index, arranging the plurality of reference data sets in the first column according to the descending / ascending order of the classification index of the plurality of reference data sets; and / or, the attribute is a classification index, arranging the plurality of reference data sets in the first column according to the descending / ascending order of the classification index of the plurality of reference data sets; and / or, the attribute is a classification index, arranging the plurality of reference data sets in the first column according to the descending / ascending order of the classification index of the plurality of reference data sets; and / or, the attribute is a classification index, arranging the plurality of reference data sets in the first column according to the descending / ascending order of the classification index of the plurality of reference data sets; and / or, the attribute is a classification index, arranging the plurality of reference data sets in the first column according to the descending / ascending order of the classification index of the plurality of reference data sets; and / or, the attribute is a classification index, arranging the plurality of reference data sets in the first column according to the descending / ascending order of the classification index of the plurality of reference data sets; and / or, the attribute is a classification index, arranging the plurality of reference data sets in the first column according to the descending / ascending order of the classification index of the plurality of reference data sets; and / or, the attribute is a classification index, arranging the plurality of reference data sets in the first column according to the descending / ascending order of the classification index of the plurality of reference data sets; and / or, the attribute is a classification index, arranging the plurality of reference data sets in the first column according to the descending / ascending order of the classification index of the plurality of reference data sets; and / or, the attribute is a classification index, arranging the plurality of reference data sets in the first column according to the descending / ascending order of the classification index of the plurality of reference data sets; and / or, the attribute is a classification index, arranging the plurality of reference data sets in the first column according to the descending / ascending order of the classification index of the plurality of reference data sets; and / or, the attribute is a classification index, arranging the plurality of reference data sets in the first column according to the descending / ascending order of the classification index of the plurality of reference data sets; and / or, the attribute is a classification index, arranging the plurality of reference data sets in the first column according to the descending / ascending order of the classification index of the plurality of reference data sets; and / or, the attribute is a classification index, arranging the plurality of reference data sets in the first column according to the descending / ascending order of the classification index of the plurality of reference data sets; and / or, the attribute is a classification index, respectively, to obtain the evaluation index of each of the plurality of reference data sets.

34. The method of data referencing of claim 33, wherein, The reference data set is a reference patent data set, and the target data set is a target patent data set; or, the data element in the reference data set is a reference patent data, and the data element in the target data set is a target patent data. The evaluation index includes market value, or technical value, or legal value. Or, The evaluation index of the reference data set is a trend feature. The trend feature includes: cross-disciplinary degree, and / or market trend degree, and / or technical trend degree. The market trend degree at least partially represents the market value of the technical category corresponding to the reference data set; and / or The technical trend degree at least partially represents / reflects the technical value of the technical category corresponding to the reference data set.

35. The method of data referencing of claim 34, wherein, The step S5 further includes the following steps: According to the evaluation index of the plurality of reference data sets, a target evaluation index is assigned to each of the plurality of target data sets; or, According to the evaluation index of the plurality of reference data sets, a target evaluation index is respectively assigned to the corresponding target data set in the plurality of target data sets; or, Based on the evaluation index of the reference data set, a target evaluation index is assigned to the target data set according to the correlation / similarity between the target data set and the reference data set; Or, According to the evaluation index of the corresponding reference data set and the target data set in the field and / or category, the target evaluation index of the target data set is determined by weighting; Or, The step of determining the attribute of the first target data set in the one or more target data sets according to the attribute of the first reference data set in the one or more reference data sets further includes: According to the evaluation index of the first reference data set and the evaluation index of the first target data set, the target evaluation index of the first target data set is determined by weighting.

36. The method of data referencing of claim 35, wherein, The step S2 further includes the following step S2-1D: According to the first standard / first rule / first algorithm, the enterprise-level data is clustered / classified / decomposed / technically decomposed to obtain a plurality of target data sets of the first enterprise and a plurality of target data sets of the second enterprise; The step A8 further includes: The plurality of reference data sets, the plurality of target data sets of the first enterprise and the plurality of target data sets of the second enterprise are respectively presented in the first column, the second column and the third column; the first column, the second column and the third column are arranged side by side / parallel along the horizontal or vertical direction; The data reference method further includes the following step A7': aligning / arranging the mutually corresponding data sets in i) the reference data set, ii) the plurality of target data sets of the first enterprise and iii) the plurality of target data sets of the second enterprise in the same row; and The correlation or similarity is: i) the intersection of the target data set and ii) the reference data set, in the target data set; or, iii) the intersection of the target data set and iii) the reference data set, in the reference data set; or, a) the size of the intersection of the target data set and the reference data set, and b) the size of the union of the target data set and the reference data set.

37. The method of data referencing of claim 36, wherein, the reference data set has a larger number of elements relative to the target data set of the corresponding field / category; or, the reference data set has stronger statistical features or representativeness relative to the target data set in the corresponding field / category; A) one or more enterprises that own the enterprise-level data belong to B) an industry that owns the industry-level data; wherein the plurality of target data sets, in classification / category, respectively one-to-one correspond to part or all of the plurality of reference data sets; the industry-level data and the enterprise-level data are patent data of the recent M years; wherein M = 2, 3, 5, 8, or 10; and / or, the industry-level data is not patent data of all countries / regions in the world, but only includes patent data of a single country or patent data of multiple countries.

38. An electronic device, comprising: a display; a processor unit; a memory; and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the processor unit, the one or more programs including instructions for performing any of the methods of claims 22-37.

39. An information processing apparatus for use in an electronic device with one or more processors, the apparatus comprising: means for performing any of the methods of claims 22-37.

40. The information processing apparatus of claim 39, further comprising means for outputting, via the display, at least one interface or at least one option configured in the electronic device.

41. A non-transitory computer-readable storage medium storing one or more programs, the one or more programs comprising instructions, which when executed by one or more processors of an electronic device, cause the electronic device to perform any of the methods of claims 22-37.

42. The non-transitory computer-readable storage medium of claim 41, the one or more programs further comprising instructions executable by the electronic device to output, via a display, at least one interface or at least one option configured in the electronic device.

43. An electronic device, comprising: a display, a processor unit, and a touch-sensitive surface unit configured to detect contact; wherein the processor unit is coupled with the display and the touch-sensitive surface unit, the processing unit being programmed to perform the steps of any of the methods of claims 22-37.

Images