Intellectual property transaction data system and processing method
Technical Field
The invention relates to the technical field of big data and artificial intelligence, in particular to an intellectual property trade data system and a processing method.
Background
201811093245.8 patent transaction method, device, computer device and storage medium, and discloses a patent transaction method, device, computer device and storage medium. The method comprises the following steps: if the patent transaction request is a patent sale request, performing identity verification based on seller information and patentee information of the patent to be traded, and if the identity verification is passed, updating the transaction state corresponding to the patent to be traded according to the legal state corresponding to the patent to be traded; if the patent transaction request is a patent purchase request, performing identity authentication on purchaser information, and if the identity authentication is passed, acquiring a transaction state corresponding to a patent to be transacted; and if the transaction state is a saleable state, acquiring seller information corresponding to the patent to be transacted, sending a patent purchase request to a client corresponding to the seller information, receiving transaction confirmation information fed back by the client, and finishing patent transaction processing based on the transaction confirmation information.
201910119262.2 intellectual property data protection, transaction and right-maintaining method and device based on block chain includes: synchronizing network states, the network states including: the number of online nodes and the information of newly added nodes; acquiring user information and receiving user requirements; according to the user requirements, selecting different functional programs of the system to execute, wherein the functional programs comprise: intellectual property data protection, intellectual property trading and right maintenance information inquiry. At present, the patents related to the collection of professional intellectual property data are fewer.
Disclosure of Invention
The invention aims to provide an intellectual property transaction data system and a processing method thereof, by which the acquisition, fusion, identification, processing, recommendation and query of intellectual property transaction data can be completed, and useful intellectual property transaction data can be provided for users in time.
An intellectual property transaction data system comprising:
the database is used for storing intellectual property trade data;
the data acquisition module is used for acquiring the intellectual property trade original data and carrying out classification and identification;
the data storage module is used for storing the classified and identified data into a database;
the data retrieval module is used for retrieving intellectual property trade data;
the user management module is used for managing user data and authority;
the acquisition comprises the steps of acquiring original data in real time by adopting a network robot and uploading the acquired original data by a user;
the classification identification is to adopt an identification algorithm to identify the original data as patent selling data, patent purchasing data and trademark purchasing data;
and the data after classified identification is stored in the database after being processed by a storage identification algorithm.
As an optimization, an intellectual property trading data system, comprising:
the information recommendation module is used for recommending transaction data according to preset keywords and attention information of the user;
and the machine learning module is used for processing the intellectual property transaction data in the database according to the deletion condition of the user on the identification error information in the database.
An intellectual property transaction data processing method, comprising:
s1, acquiring first original data in real time through a preset network robot, and acquiring second original data through uploading by a user;
s2, classifying the first original data and the second original data through a keyword filtering algorithm to obtain patent selling data, patent purchasing data and trademark purchasing data;
and S3, storing the patent selling data, the patent purchasing data and the trademark purchasing data into a database after being processed by a warehousing identification algorithm.
As optimization, the warehousing identification algorithm comprises:
s31, for the patent selling data, firstly, judging whether the database has the existing patent selling data which simultaneously satisfies the patent numbers and has the same contact way; if not, storing the data into a database; if yes, existing patent selling data with the same patent number and the same contact way in the database is updated;
s32, for the patent purchase data, firstly judging whether existing patent purchase data with the same contact way exists; if not, storing the patent purchase data into a database; if yes, judging the similarity psi between the existing patent purchase data and the patent purchase data;
if the similarity psi is larger than a preset threshold value omega, updating the existing patent purchase data; otherwise, storing the patent purchase data into a database;
s33, for the trademark purchasing data, firstly judging whether existing trademark purchasing data with the same contact way exists; if not, storing the trademark purchase data into a database; if yes, judging the similarity psi between the existing trademark purchase data and the trademark purchase data;
if the similarity psi is larger than a preset threshold value omega, updating the purchase data of the existing trademark; otherwise, store the trade mark purchase data in the database.
As an optimization, the warehousing identification algorithm comprises:
s31, for the patent selling data, firstly, judging whether the database has the existing patent selling data which simultaneously satisfies the patent numbers and has the same contact way; if not, storing the data into a database; if yes, judging whether the data source of the patent selling data, the patent number and the existing patent selling data with the same contact way is from the first original data or the second original data;
if the patent selling data is from the first original data, the existing patent selling data with the same patent number and the same contact way is from the second original data; storing no patent selling data, otherwise updating the existing patent selling data with the same patent number and the same contact way in the database;
s32, for the patent/trademark purchase data, firstly judging whether existing patent/trademark purchase data with the same contact way exists; if not, storing the patent/trademark purchase data into a database; if yes, judging the similarity psi between the existing patent/trademark purchase data and the patent/trademark purchase data;
if the similarity psi is not greater than the preset threshold omega, storing the patent/trademark purchase data into a database;
if the similarity psi is larger than a preset threshold value omega, judging whether the data source of the patent/trademark purchase data and the data source of the existing patent/trademark purchase data with the similarity psi larger than the preset threshold value omega come from the first original data or the second original data;
if the patent/trademark purchase data is from the first original data, the existing patent/trademark purchase data with the similarity psi larger than the preset threshold value omega is from the second original data; the patent/trademark purchase data is not stored, otherwise the existing patent/trademark purchase data having the similarity ψ larger than the preset threshold ω in the database is updated.
As optimization, the intellectual property trade data processing method comprises the following steps:
processing the intellectual property transaction data existing in the database according to the deletion condition of the user on the identification error information in the database;
the processing database has intellectual property trade data as follows:
deleting the identification error information in the database by the authorized user to obtain identification error information data;
performing word segmentation operation on each piece of identification error information data, each piece of patent purchase data and each piece of trademark purchase data;
calculating related data in patent purchase data and trademark purchase data according to a word segmentation of a certain piece of recognition error information data;
the related data is data of all words after a certain piece of recognition error information data is segmented;
and calculating the similarity α of the related data and a certain piece of identification error information data one by one, and marking the related data as error information if the similarity α is greater than a threshold β.
As optimization, the intellectual property trade data processing method comprises the following steps:
recommending transaction data through a recommendation algorithm according to preset keywords and attention information of a user;
the recommendation algorithm is as follows:
according to preset keywords of a user, obtaining recommended patent selling data or recommended patent purchasing data or recommended trademark purchasing data with the preset keywords in a database;
calculating the recommendation degree omicron of each recommended patent sale data according to the patent sale data which are concerned by the user and contain preset keywords; and determining the recommended sequence according to the recommendation degree.
As optimization, a processing method of intellectual property trade data, wherein the calculation method of the recommendation degree o is as follows:
acquiring a patent name and a patent classification number of patent sale data which are concerned by a user and contain preset keywords;
calculating the similarity gamma between the recommended patent selling data and the patent name of the patent selling data which contains preset keywords and is concerned by the user;
calculating the proximity delta of the recommended patent selling data and the patent selling data patent classification number which contains preset keywords and is concerned by the user;
recommendation degree omicron = γ + δ;
the patent classification number includes a principal classification number and a classification number;
proximity δ = ε ζ + θ ν/n;
epsilon is the weight of the primary classification number,
zeta is that whether the recommended patent selling data is the same as the main classification number of the patent selling data which contains the preset keywords and is concerned by the user, wherein the same is 1, and the different is 0;
theta is the weight of the classification number and,
ν is the same number of classification numbers of the recommended patent selling data and the patent selling data which contains preset keywords and is concerned by the user;
n=min(κ,λ);
kappa is the number of classification numbers of recommended patent selling data;
and lambda is the number of the patent sale data classification numbers which are concerned by the user and contain preset keywords.
The invention relates to an intellectual property trade data system and a processing method, which collect original data in two ways; after the data are classified through a keyword filtering algorithm, different source data are fused into a database through a warehousing identification algorithm;
meanwhile, through machine learning, the user deletes the identification error information in the database, and optimizes the error data of the existing intellectual property trade data in the database; and recommending the latest patent transaction data to the user through a recommendation algorithm. The intellectual property transaction data system disclosed by the invention has the advantages that the data with various sources are intelligently fused, the maintenance and the processing of the data are more intelligent, and meanwhile, the useful intellectual property transaction data can be intelligently recommended to users.
Drawings
The invention is described in detail below with reference to the drawings and the detailed description;
FIG. 1 is a schematic flow chart according to embodiment 1 of the present invention;
FIG. 2 is a schematic flow chart according to embodiment 2 of the present invention;
fig. 3 is a schematic diagram of a warehousing identification algorithm according to embodiment 2 of the present invention;
fig. 4 is a schematic diagram of a warehousing identification algorithm according to embodiment 3 of the present invention;
FIG. 5 is a diagram of a publishing interface according to embodiment 4 of the present invention;
FIG. 6 is a search interface diagram according to embodiment 4 of the present invention;
FIG. 7 is a diagram of a recommendation interface according to embodiment 4 of the present invention;
fig. 8 is a statistical interface diagram according to embodiment 4 of the present invention.
Detailed Description
The following examples are intended to further illustrate the present invention and should not be construed as limiting the scope of the invention, and other insubstantial modifications and adaptations of the invention by those skilled in the art based on the teachings herein are intended to be covered thereby.
Example 1: as shown in fig. 1, an intellectual property transaction data system includes:
the database is used for storing intellectual property trade data;
the data acquisition module is used for acquiring the intellectual property trade original data and carrying out classification and identification; the acquisition comprises the steps of acquiring original data in real time by adopting a network robot and uploading the acquired original data by a user; acquiring original data in real time by adopting a network robot, wherein the method comprises the steps of acquiring information in a QQ group in real time by adopting a QQ robot and legally acquiring open information by adopting a bent-path crawler;
classifying and identifying, namely identifying the original data as patent selling data, patent purchasing data and trademark purchasing data by adopting an identification algorithm;
the data storage module is used for storing the classified and identified data into a database; storing the data into a database, processing the classified and recognized data through a storage recognition algorithm, and storing the data into the database;
the data retrieval module is used for retrieving intellectual property trade data;
the user management module is used for managing user data and authority;
the information recommendation module is used for recommending transaction data according to preset keywords and attention information of the user;
and the machine learning module is used for processing the intellectual property transaction data in the database according to the deletion condition of the user on the identification error information in the database.
Example 2: as shown in fig. 2, the intellectual property trade data processing method includes:
s1, acquiring first original data in real time through a preset network robot, and acquiring second original data through uploading by a user;
s2, classifying the first original data and the second original data through a keyword filtering algorithm to obtain patent selling data, patent purchasing data and trademark purchasing data;
s3, storing the patent selling data, the patent purchasing data and the trademark purchasing data into a database after being processed by a warehousing identification algorithm;
s4, processing the intellectual property trade data in the database according to the deletion condition of the user to the identification error information in the database;
and S5, recommending the transaction data through a recommendation algorithm according to the preset keywords and the attention information of the user.
The warehousing identification algorithm is shown in fig. 3 and includes:
s31, for the patent selling data, firstly, judging whether the database has the existing patent selling data which simultaneously satisfies the patent numbers and has the same contact way; if not, storing the data into a database; if yes, existing patent selling data with the same patent number and the same contact way in the database is updated;
s32, for the patent purchase data, firstly judging whether existing patent purchase data with the same contact way exists; if not, storing the patent purchase data into a database; if yes, judging the similarity psi between the existing patent purchase data and the patent purchase data;
if the similarity psi is larger than a preset threshold value omega, updating the existing patent purchase data; otherwise, storing the patent purchase data into a database;
s33, for the trademark purchasing data, firstly judging whether existing trademark purchasing data with the same contact way exists; if not, storing the trademark purchase data into a database; if yes, judging the similarity psi between the existing trademark purchase data and the trademark purchase data;
if the similarity psi is larger than a preset threshold value omega, updating the purchase data of the existing trademark; otherwise, store the trade mark purchase data in the database.
The similarity ψ may adopt a cosine similarity algorithm, a simple common word algorithm, an edit distance algorithm, an euclidean/manhattan distance algorithm, a Jaccard similarity coefficient algorithm, or the like.
The method for processing the existing intellectual property transaction data in the database comprises the following steps:
deleting the identification error information in the database by the authorized user to obtain identification error information data;
performing word segmentation operation on each piece of identification error information data, each piece of patent purchase data and each piece of trademark purchase data;
calculating related data in the patent purchase data and the trademark purchase data according to the word segmentation of a certain piece of recognition error information data A; the related data is data of all words after a certain piece of recognition error information data is segmented; the related data may or may not be one or more;
the similarity α of the related data to a certain piece of the identification error information data A is calculated one by one, and if the similarity α is larger than the threshold β, the related data is marked as error information.
The recommendation algorithm is as follows:
according to preset keywords of a user, obtaining recommended patent selling data or recommended patent purchasing data or recommended trademark purchasing data with the preset keywords in a database;
calculating the recommendation degree omicron of each recommended patent sale data according to the patent sale data which are concerned by the user and contain preset keywords; and determining the recommendation sequence of the patent sale data according to the recommendation degree.
The calculation method of the recommendation degree o is as follows:
acquiring a patent name and a patent classification number of patent sale data which are concerned by a user and contain preset keywords;
calculating the similarity gamma between the recommended patent selling data and the patent name of the patent selling data which contains preset keywords and is concerned by the user;
calculating the proximity delta of the recommended patent selling data and the patent selling data patent classification number which contains preset keywords and is concerned by the user;
recommendation degree omicron = γ + δ;
the patent classification number includes a principal classification number and a classification number;
proximity δ = ε ζ + θ ν/n; e.g., e =0.6, θ = 0.4;
epsilon is the weight of the primary classification number,
zeta is that whether the recommended patent selling data is the same as the main classification number of the patent selling data which contains the preset keywords and is concerned by the user, wherein the same is 1, and the different is 0;
theta is the weight of the classification number and,
ν is the same number of classification numbers of the recommended patent selling data and the patent selling data which contains preset keywords and is concerned by the user;
n=min(κ,λ);
kappa is the number of classification numbers of recommended patent selling data;
and lambda is the number of the patent sale data classification numbers which are concerned by the user and contain preset keywords.
Example 3: the intellectual property transaction data processing method in embodiment 2 is different in that the warehousing identification algorithm is shown in fig. 4 and includes:
s31, for the patent selling data, firstly, judging whether the database has the existing patent selling data which simultaneously satisfies the patent numbers and has the same contact way; if not, storing the data into a database; if yes, judging whether the data source of the patent selling data, the patent number and the existing patent selling data with the same contact way is from the first original data or the second original data;
if the patent selling data is from the first original data, the existing patent selling data with the same patent number and the same contact way is from the second original data; storing no patent selling data, otherwise updating the existing patent selling data with the same patent number and the same contact way in the database;
s32, for the patent/trademark purchase data, firstly judging whether existing patent/trademark purchase data with the same contact way exists; if not, storing the patent/trademark purchase data into a database; if yes, judging the similarity psi between the existing patent/trademark purchase data and the patent/trademark purchase data;
if the similarity psi is not greater than the preset threshold omega, storing the patent/trademark purchase data into a database;
if the similarity psi is larger than a preset threshold value omega, judging whether the data source of the patent/trademark purchase data and the data source of the existing patent/trademark purchase data with the similarity psi larger than the preset threshold value omega come from the first original data or the second original data;
if the patent/trademark purchase data is from the first original data, the existing patent/trademark purchase data with the similarity psi larger than the preset threshold value omega is from the second original data; the patent/trademark purchase data is not stored, otherwise the existing patent/trademark purchase data having the similarity ψ larger than the preset threshold ω in the database is updated.
Example 4: an intellectual property transaction data system is currently developed, comprising:
the database is used for storing intellectual property transaction data and is arranged at the cloud server side;
the data acquisition module acquires information in the QQ group in real time by adopting a QQ robot, and acquires original data through user uploading to finish data acquisition; classifying through a keyword filtering algorithm to obtain patent selling data, patent purchasing data and trademark purchasing data;
the user uploads the data file directly or uploads the data file to be published after automatically identifying the pasted content like the publishing interface in the figure 5;
a data warehousing module, configured to store the classified and identified data into the database after being processed by the warehousing identification algorithm in embodiment 2 or 3, where the statistics of the warehousing identification algorithm update data, i.e., the number of modified pieces, is shown in fig. 8, and the data statistics directly stored in the database, i.e., the number of newly added pieces; the collected patent selling data is corrected through a patent database interface;
the data retrieval module is used for intellectual property transaction data retrieval, comprises functions of screening, selecting and the like, and is used for visual retrieval operation of customers like the retrieval interface shown in FIG. 6;
the user management module is used for managing user data and authority, and comprises the levels of users, and the viewing ranges and functions of all levels;
the information recommendation module is used for recommending transaction data according to preset keywords and attention information of the user through a recommendation algorithm similar to that in embodiment 2; as shown in fig. 7, the recommendation interface of the patent information is displayed by setting a preset tag, namely a keyword;
and the machine learning module is used for processing the intellectual property transaction data in the database according to the deletion condition of the user on the identification error information in the database.
While the invention has been described in connection with specific embodiments thereof, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.