CN117519702A - Search page design method and system based on low code collocation - Google Patents
Search page design method and system based on low code collocation Download PDFInfo
- Publication number
- CN117519702A CN117519702A CN202311850582.8A CN202311850582A CN117519702A CN 117519702 A CN117519702 A CN 117519702A CN 202311850582 A CN202311850582 A CN 202311850582A CN 117519702 A CN117519702 A CN 117519702A
- Authority
- CN
- China
- Prior art keywords
- search
- query
- operator
- recommendation
- database
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 44
- 238000013461 design Methods 0.000 title claims abstract description 14
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 30
- 238000012545 processing Methods 0.000 claims abstract description 6
- 230000006399 behavior Effects 0.000 claims description 21
- 238000012549 training Methods 0.000 claims description 13
- 238000003066 decision tree Methods 0.000 claims description 10
- 230000006870 function Effects 0.000 claims description 9
- 239000011159 matrix material Substances 0.000 claims description 9
- 238000004590 computer program Methods 0.000 claims description 5
- 230000008447 perception Effects 0.000 claims description 5
- 230000007547 defect Effects 0.000 claims description 3
- 238000012216 screening Methods 0.000 claims description 3
- 230000009466 transformation Effects 0.000 claims description 3
- 238000011161 development Methods 0.000 description 4
- 239000000284 extract Substances 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013499 data model Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000007637 random forest analysis Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F8/00—Arrangements for software engineering
- G06F8/30—Creation or generation of source code
- G06F8/38—Creation or generation of source code for implementing user interfaces
Landscapes
- Engineering & Computer Science (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention provides a search page design method and a system based on low code collocation, which relate to the technical field of data processing, and provides a search page designer and a search model, wherein a search column name list is provided for the search page designer based on the search model, a customized search component is connected with the search model, column names are selected through dragging, and a search field is created in the customized search component; selecting at least one query operator from a query operator list adapting to the search field, binding with the search field to form a query condition element, and combining the query condition element by using a rule expression in a query rule configurator to form a query rule; providing a search executor, analyzing a query operator into an operator algorithm unit queue based on a query rule, recombining the operator algorithm unit queue into a database query condition by combining a logic operator, forming a database query statement by combining a table of a database, and querying in the database.
Description
Technical Field
The present invention relates to data processing technologies, and in particular, to a search page design method and system based on low code collocation.
Background
Along with the increasing of the user quantity, the data quantity stored in the service system is also increasing, based on which the data required by the user is rapidly searched from the massive data, the user is more and more important to search, and more abundant and variable search conditions are required to participate in the search, if a configuration method of a group of search pages can be provided, the configuration is not required to be realized by a developer through coding, the configuration is effective in real time through dragging, the reissue is not required, the usability of an application program is improved, and the development cost is greatly reduced.
CN202310704040.3 discloses a query method, device, electronic equipment and storage medium based on a configurable search component, the method comprising: acquiring combined search conditions through a front-end search component; the front-end searching component transmits the combined searching condition to the back-end searching component; the back-end searching component performs data query on the combined searching conditions and transmits the obtained query results to the front-end searching component; the front-end search component presents the query results. Therefore, the front-end searching assembly and the back-end searching assembly in the searching assembly are used for inquiring data, different functional modules can be directly adapted, the expansibility and mobility of the searching function are improved, and the development is not required again, so that the workload of subsequent development and dimension is reduced. Meanwhile, query condition item logic can be automatically combined among conditions in the combined search conditions, so that complex logic judgment of a user is met.
CN202111413576.7 discloses a search method, device, equipment and computer readable storage medium based on search server, by obtaining target assignment and search query grammar template; determining a search query grammar statement according to the search query grammar template and the target assignment; transmitting the search query grammar statement to a search server; and receiving the search result returned by the search server. The preset search query grammar template is equivalent to the search requirement which is presupposed, and aiming at the compatibility problem of the ES of different versions and the local framework and the ES, only the search query grammar template is required to be changed in a targeted manner, and the personnel actually executing the query do not need to further learn the advanced client and version writing difference of the ES, so that the learning cost is greatly reduced, and simultaneously, aiming at the new requirement and version change, only the search query grammar template is required to be directly modified in the text, thereby being quick and convenient and greatly improving the usability and maintainability of codes.
In summary, in the prior art, the front end and the rear end interact through modularized search or through grammar templates, the core still uses fixed search conditions, and the search function interface corresponding to the core is provided for the rear end service, so that flexibility is lacking, in order to enable a developer to take effect in real time without coding, and reissue through dragging configuration, the usability of an application program is not required, the development cost is greatly reduced, and the application of the invention at least can solve part of the problems in the prior art.
Disclosure of Invention
The embodiment of the invention provides a search page design method and a search page design system based on low code collocation, which at least can solve part of problems in the prior art.
In a first aspect of an embodiment of the present invention,
providing a search page designer and a search model, providing a search column name list for the search page designer based on the search model preset by a table of a database, providing a customized search component by the search page designer, establishing connection with the search model, selecting column names in the search column name list by dragging, and creating a search field in the customized search component;
at least one query operator is selected from a query operator list adapting to the search field and is bound with the search field to form a query condition element, and the query condition element is combined through a logic operator by using a rule expression in a query rule configurator preset with a logic operator to form a query rule;
providing a search executor comprising an operator algorithm unit, analyzing the query operator in the query condition element into an operator algorithm unit queue based on the query rule acquired from the search page designer, combining a logic operator of the query rule, recombining the operator algorithm unit queue into a database query condition, combining a table of a database corresponding to the search model with the database query condition to form a database query statement, and querying in the database based on the database query statement.
In an alternative embodiment of the present invention,
creating a search field in the customized search component by dragging and selecting a column name in the search column name list, further comprising:
based on the field names collected from the database, a corpus is established, and the corpus is used for training a search intention analyzer;
collecting historical search behavior data of a user in the search page designer, wherein the search intention analyzer learns the user behavior through the historical search behavior data, and the historical search behavior data comprises search preference, a common model, a common field and a use frequency of the user;
setting a search intention text box in the search page designer, inputting a search intention expression language in the search intention text box by a user, analyzing the search intention expression language by the search intention analyzer, extracting keywords and field information, and screening and sequentially presenting user intention fields in combination with the user behaviors.
In an alternative embodiment of the present invention,
training a search intent analyzer using the corpus includes:
setting a sensing parameter dynamically adjusted according to a context, wherein the sensing parameter is formed by vector connection of an input vector of a target word and an input vector of the whole context through a Concat function, performing linear transformation by using a parameter matrix for adjusting the influence of the context, setting a bias term by considering the mutual influence of the context, and limiting the sensing parameter between 0 and 1 through a sigmoid function, wherein the formula is as follows:
；
wherein,alpha representsThe sensing parameters are used to determine the sensing parameters,W α the parameter matrix is represented by a matrix of parameters,v wt an input vector representing the target word,v context vector representation representing the entire context, concatv wt ,v context ) Representing the target wordwtIs concatenated with the vector representation of the entire context,b α representing the bias term.
In an alternative embodiment of the present invention,
the method further comprises the steps of:
the search intention analyzer uses a search language processing model to calculate intention probability by combining the perception parameters, and the formula is as follows:
；
wherein,wcthe context-based words are represented by the terms "a",wtthe target word is represented by a word of interest,P(wc|wt) Representing target wordswtUnder the conditions of appearance, the context wordswcThe probability of the occurrence of the presence of a defect,αrepresenting the perceived parameter(s),v' wc representing context wordswcIs used to determine the output vector of (a),v wt an input vector representing the target word,cnrepresents one of the context words in the whole,v' cn representing context wordscnIs provided.
In an alternative embodiment of the present invention,
selecting at least one query operator from a list of query operators adapting to the search field, wherein the at least one query operator is bound with the search field, and the method further comprises:
acquiring historical binding data of a user through the search list name list and the query operator list, extracting historical binding characteristics, and obtaining a comprehensive recommendation result of the query operator through comprehensive recommendation of the first operator recommendation model and the second operator recommendation model:
according to the historical binding characteristics, using a field name and a field type as input parameters of a first operator recommendation model, randomly extracting samples to form a characteristic subset through the first operator recommendation model, establishing a decision tree, recursively carrying out node splitting to obtain a decision result, integrating a plurality of decision trees, and determining a first recommendation result according to the decision results;
according to the history binding characteristics, taking a target data point formed by a field name and a field type as an input parameter of a second operator recommendation model, calculating the distance between the target data point and other data points through the second operator recommendation model, obtaining a data point list based on the sequential arrangement from small to large, selecting a plurality of data points with the distance smaller than a preset distance threshold value from the data point list as threshold neighbors of the target data point, counting the number of query operators of the threshold neighbors, and taking the query operator with the largest number of query operators as a second recommendation result;
and determining a first recommendation weight corresponding to the first recommendation result and a second recommendation weight corresponding to the second recommendation result by using an information entropy method when comprehensive recommendation is performed, and determining the comprehensive recommendation result by combining the first recommendation result and the second recommendation result.
In an alternative embodiment of the present invention,
when the comprehensive recommendation is performed, determining a first recommendation weight corresponding to the first recommendation result and a second recommendation weight corresponding to the second recommendation result by using an information entropy method, and determining the comprehensive recommendation result by combining the first recommendation result and the second recommendation result comprises:
the weight is calculated by using an information entropy method, and the information entropy is expressed as follows:
；
wherein,Hthe entropy of the information is represented and,x i representing the first of the modelsiThe result of the recommendation is referred to as,P(x i ) Expressed in the model asiProbability of individual recommendation results;
the weight is calculated as follows:
；
wherein,ωthe weight is represented by a weight that,Hthe entropy of the information is represented and,ea base representing natural logarithms;
the formula of the comprehensive recommendation result is as follows:
；
wherein,Tthe result of the integrated recommendation is indicated,ω 1 a first recommendation weight is indicated and a first recommendation weight is indicated,ω 2 a second recommendation weight is indicated as such,R 1 a first recommendation result is indicated and a first recommendation result is indicated,R 2 and representing a second recommendation result.
In a second aspect of an embodiment of the present invention,
there is provided a search page design system based on low code configurational, comprising:
the first unit is used for providing a search page designer and a search model, providing a search column name list for the search page designer based on the search model preset by a table of a database, providing a customized search component by the search page designer, establishing connection with the search model, and creating a search field in the customized search component by dragging and selecting a column name in the search column name list;
a second unit, configured to select at least one query operator from a query operator list adapted to the search field, bind with the search field, and form a query condition element, where in a query rule configurator preset with a logic operator, a rule expression is used to combine the query condition element through the logic operator, so as to form a query rule;
and the third unit is used for providing a search executor comprising an operator algorithm unit, analyzing the query operator in the query condition element into an operator algorithm unit queue based on the query rule acquired from the search page designer, combining a logic operator of the query rule, reconstructing the operator algorithm unit queue into a database query condition, combining a table of a database corresponding to the search model with the database query condition to form a database query statement, and querying in the database based on the database query statement.
In a third aspect of an embodiment of the present invention,
there is provided an electronic device including:
a processor;
a memory for storing processor-executable instructions;
wherein the processor is configured to invoke the instructions stored in the memory to perform the method described previously.
In a fourth aspect of an embodiment of the present invention,
there is provided a computer readable storage medium having stored thereon computer program instructions which, when executed by a processor, implement the method as described above.
In the embodiment of the invention, a user can easily customize a search interface through a search page designer, select a required search field according to service requirements, and drag and select a column name so that customization becomes visual and user-friendly; through training of the search intention analyzer, the system can more intelligently understand the intention of the user, extract keywords and field information, and set search conditions more conveniently; by collecting field names from a database, establishing a corpus, training a search intention analyzer and the like, the system realizes automatic field recommendation and search intention analysis, reduces the burden of a user for configuring search conditions, and improves the usability of the system; the user can freely select the search field and the query operator through dragging and configuration operation, and construct a complex query rule, so that the customized search page has flexibility and customization, and the query requirements of different users are met; by acquiring the query rules from the search page designer and analyzing the query condition elements into an operator algorithm unit queue, the system can construct flexible and intelligent query sentences according to user input, so that the system can be more suitable for the diversified query requirements of users instead of fixed predefined query; because of the flexibility and intelligence of the operator algorithm unit queue, the system can better adapt to structural changes of a database table or different query scenes of users, high adaptability is maintained, the users only need to configure query rules in a search page designer, the system can automatically analyze the query rules into database query sentences, the burden of manually writing complex query sentences by the users is reduced, and user experience is improved.
Drawings
FIG. 1 is a flow chart of a search page design method based on low code configuration according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of a search page design system based on low code configuration according to an embodiment of the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The technical scheme of the invention is described in detail below by specific examples. The following embodiments may be combined with each other, and some embodiments may not be repeated for the same or similar concepts or processes.
Fig. 1 is a flow chart of a search page design method based on low code configuration according to an embodiment of the present invention, as shown in fig. 1, the method includes:
s101, providing a search page designer and a search model, providing a search column name list for the search page designer based on the search model preset by a table of a database, providing a customized search component by the search page designer, establishing connection with the search model, selecting column names in the search column name list through dragging, and creating a search field in the customized search component;
the search model specifically refers to an abstract data model for describing searchable data objects in the system and attributes of the data objects, each model corresponds to a table in a database, each table comprises a plurality of columns, and column names of the columns form a search column name list;
the method comprises the steps that a search model corresponds to a table of a database one by one, a column name list in the database table is extracted based on the search model, the column name list is used for displaying fields available for searching in a search page designer, each field in the column name list is bound with the table of the database, and corresponding fields are displayed on a customized search component in a dragging or choosing mode, wherein the fields are search fields;
in this embodiment, the user can easily customize the search interface by searching the page designer, select a required search field according to the service requirement, drag and select the column name to make the customization intuitive and user-friendly, and the search model provides abstraction of the database table, so that accuracy and consistency of the front search field are ensured, and possibility of manual input errors is avoided.
In an alternative embodiment, creating a search field in the customized search component by dragging and selecting a column name in the search column name list, further comprises:
based on the field names collected from the database, a corpus is established, and the corpus is used for training a search intention analyzer;
collecting historical search behavior data of a user in the search page designer, wherein the search intention analyzer learns the user behavior through the historical search behavior data, and the historical search behavior data comprises search preference, a common model, a common field and a use frequency of the user;
setting a search intention text box in the search page designer, inputting a search intention expression language in the search intention text box by a user, analyzing the search intention expression language by the search intention analyzer, extracting keywords and field information, and screening and sequentially presenting user intention fields in combination with the user behaviors.
The search intention analyzer specifically refers to a module for analyzing a search intention expression language input by a user in a search page, extracting key words and field information in the search intention expression language, and learning according to historical search behavior data of the user so as to understand and explain the search intention of the user;
collecting field names from a database, establishing a corpus containing vocabulary and context information related to the fields, training a search intention analyzer by using the established corpus to understand semantic relations among the field names, collecting historical search behavior data of a user in a search page designer, including information such as search preference, common model, common field, use frequency and the like, and understanding the search mode, preference and common field of the user by learning the historical search behavior data;
setting a search intention text box in a search page designer, allowing a user to input a natural language expression of a search intention, and interacting with a search intention analyzer through the search intention text box; when a user inputs a search intention expression language in a search intention text box, a search intention analyzer analyzes the input, extracts keywords and field information, and combines the historical search behaviors of the user, and the search intention analyzer screens and sequentially presents user intention fields;
in the embodiment, through training of the search intention analyzer, the system can more intelligently understand the intention of the user, extract keywords and field information, and set search conditions more conveniently; by collecting field names from a database, establishing a corpus, training a search intention analyzer and the like, the system realizes automatic field recommendation and search intention analysis, reduces the burden of a user for configuring search conditions, and improves the usability of the system;
in an alternative embodiment, training the search intent analyzer using the corpus includes:
setting a sensing parameter dynamically adjusted according to a context, wherein the sensing parameter is formed by vector connection of an input vector of a target word and an input vector of the whole context through a Concat function, performing linear transformation by using a parameter matrix for adjusting the influence of the context, setting a bias term by considering the mutual influence of the context, and limiting the sensing parameter between 0 and 1 through a sigmoid function, wherein the formula is as follows:
；
wherein,alpha representsThe sensing parameters are used to determine the sensing parameters,W α the parameter matrix is represented by a matrix of parameters,v wt an input vector representing the target word,v context vector representation representing the entire context, concatv wt ,v context ) Representing the target wordwtIs concatenated with the vector representation of the entire context,b α representing the bias term.
In an alternative embodiment, the method further comprises:
the search intention analyzer uses a search language processing model to calculate intention probability by combining the perception parameters, and the formula is as follows:
；
wherein,wcthe context-based words are represented by the terms "a",wtthe target word is represented by a word of interest,P(wc|wt) Representing target wordswtUnder the conditions of appearance, the context wordswcThe probability of the occurrence of the presence of a defect,αrepresenting the perceived parameter(s),v' wc representing context wordswcIs used to determine the output vector of (a),v wt an input vector representing the target word,cnrepresents one of the context words in the whole,v' cn representing context wordscnIs provided.
In the embodiment, a corpus is built based on collected field names, semantic analysis and intention inference are carried out on search intention input by a user through a search intention analyzer, a more intelligent field creation is realized, the search intention analyzer learns search preference, a common model, a field and use frequency of the user through collecting historical search behavior data of the user, so that personalized search suggestions are provided, perception parameters are introduced to carry out context dynamic adjustment, parameters of search intention analysis can be adaptively adjusted according to context information, flexibility and adaptability of the model are improved, conditional probability of each word in the context is calculated by utilizing the model and the perception parameters obtained through training, the relation among each word in the search intention is more accurately captured, search accuracy is improved, the user can enjoy more flexible, intelligent and personalized search experience, and search efficiency and accuracy are improved.
S102, selecting at least one query operator from a query operator list adapting to the search field, binding with the search field to form a query condition element, and combining the query condition element through a logic operator by using a rule expression in a query rule configurator preset with a logic operator to form a query rule;
the query operators specifically refer to operators or functions for constructing query conditions, are used for describing the relation between a search field and a specific value, define the properties of the search conditions, such as equal to, greater than, less than, and the like, and each query operator is generally bound with the search field to represent the specific operation on the field;
the query condition element sub-specifically refers to a single condition unit composed of a search field and a query operator bound with the search field, and represents a specific condition set by a user in customized search, such as 'name, inclusion', namely a query condition element sub-;
the query rule specifically refers to a logic expression formed by combining a plurality of query condition elements through a logic operator, wherein the logic operator comprises AND (AND), OR (OR) AND the like, AND is used for connecting different query condition elements to form a complex query rule, for example, "OR (name. Containing, AND (name. Equals, time. Less than))" is a query rule, AND the meaning of the query rule is: the name contains a result set of search value 1 or a result set of name equal to search value 1 and time less than search value 2.
For each search field, selecting at least one query operator from a preset query operator list, for example, for a 'name' field, an equal operator and an equal operator can be selected and contained; each search field is combined with a selected query operator to form a query condition element, and in the query rule configurator, a user can combine a plurality of query condition element elements through a preset logical operator in a dragging or selecting mode to form a complex query rule.
In the embodiment, the user can freely select the search field and the query operator through dragging and configuration operation, and construct a complex query rule, so that the customized search page has flexibility and customization, and the query requirements of different users are met; the dragging and configuration modes are used, so that the searching setting becomes visual and easy to operate, the learning cost of a user is reduced, and the usability of the system is improved; the system can analyze the query rule configured by the user, convert the query rule into a query condition element sub-and combine the query condition element sub-according to a logic operator;
in an alternative embodiment, at least one query operator is selected from a list of query operators adapted to the search field to bind to the search field, further comprising:
acquiring historical binding data of a user through the search list name list and the query operator list, extracting historical binding characteristics, and obtaining a comprehensive recommendation result of the query operator through comprehensive recommendation of the first operator recommendation model and the second operator recommendation model:
according to the historical binding characteristics, using a field name and a field type as input parameters of a first operator recommendation model, randomly extracting samples to form a characteristic subset through the first operator recommendation model, establishing a decision tree, recursively carrying out node splitting to obtain a decision result, integrating a plurality of decision trees, and determining a first recommendation result according to the decision results;
according to the history binding characteristics, taking a target data point formed by a field name and a field type as an input parameter of a second operator recommendation model, calculating the distance between the target data point and other data points through the second operator recommendation model, obtaining a data point list based on the sequential arrangement from small to large, selecting a plurality of data points with the distance smaller than a preset distance threshold value from the data point list as threshold neighbors of the target data point, counting the number of query operators of the threshold neighbors, and taking the query operator with the largest number of query operators as a second recommendation result;
and determining a first recommendation weight corresponding to the first recommendation result and a second recommendation weight corresponding to the second recommendation result by using an information entropy method when comprehensive recommendation is performed, and determining the comprehensive recommendation result by combining the first recommendation result and the second recommendation result.
The first operator recommendation model is specifically constructed based on a random forest tree model, and the second operator recommendation model is specifically constructed based on a K nearest neighbor algorithm;
extracting historical binding characteristics related to field names and field types by using the field names and the field types as input parameters and combining the historical binding data, forming a sample set by using the historical binding characteristics, randomly extracting samples to form a characteristic subset, recursively carrying out node splitting, establishing a decision tree, obtaining a decision result according to the node splitting condition of the decision tree, integrating a plurality of decision trees based on multiple training, and determining a result given by a first operator recommendation model according to the results of the plurality of decision trees;
forming a target data point by using a field name and a field type as input parameters, extracting historical binding characteristics related to the field name and the field type by using historical binding data, calculating the distance between the target data point and other data points by using a second operator recommendation model, sequentially arranging the data points from small to large based on the distance to obtain a data point list, selecting the data points with the distance smaller than a preset distance threshold value, forming a threshold neighbor, counting the number of query operators of the threshold neighbor, and taking the query operator with the largest number of query operators as a result given by the second operator recommendation model;
for a first recommendation result, calculating the frequency distribution of the first recommendation result in the history binding data to obtain a first information entropy; for the second recommendation result, the frequency distribution of the second recommendation result in the history binding data is calculated to obtain a second information entropy; calculating normalized information entropy as the weights of the first recommendation result and the second recommendation result; and linearly weighting the first recommendation result and the second recommendation result according to the weight calculated by the information entropy method, and comprehensively obtaining a final query operator comprehensive recommendation result.
In an alternative embodiment, determining, using an entropy method, a first recommendation weight corresponding to the first recommendation result and a second recommendation weight corresponding to the second recommendation result when the integrated recommendation is performed, and determining, in combination with the first recommendation result and the second recommendation result, the integrated recommendation result includes:
the weight is calculated by using an information entropy method, and the information entropy is expressed as follows:
；
wherein,Hthe entropy of the information is represented and,x i representing the first of the modelsiThe result of the recommendation is referred to as,P(x i ) Expressed in the model asiProbability of individual recommendation results;
the weight is calculated as follows:
；
wherein,ωthe weight is represented by a weight that,Hthe entropy of the information is represented and,ea base representing natural logarithms;
the formula of the comprehensive recommendation result is as follows:
；
wherein,Tthe result of the integrated recommendation is indicated,ω 1 a first recommendation weight is indicated and a first recommendation weight is indicated,ω 2 a second recommendation weight is indicated as such,R 1 a first recommendation result is indicated and a first recommendation result is indicated,R 2 and representing a second recommendation result.
In the embodiment, through analyzing historical binding data of the user, the first operator recommendation model can recommend the most probable query operator for the user according to the field name and the field type, so that personalized recommendation is realized; the second operator recommendation model calculates the distance between the target data point and other data points by using the distance measurement, so that the data point with the closer distance is found to serve as the neighbor of the target data point, and more relevant query operator selections are provided for the user through neighborhood analysis; the weight is calculated by using an information entropy method, the distribution situation of different query operators in the historical binding data is fully considered, the preference of the historical behavior of the user can be reflected more accurately, and the intelligent degree of recommendation is improved; the first operator recommendation and the second operator recommendation are combined in a linear weighting mode, so that a comprehensive and comprehensive query operator recommendation result is obtained, the intelligent level of the system is improved, and more expected recommendation is provided for users.
S103, providing a search executor comprising an operator algorithm unit, analyzing the query operator in the query condition element into an operator algorithm unit queue based on the query rule acquired from the search page designer, combining a logic operator of the query rule, recombining the operator algorithm unit queue into a database query condition, combining a table of a database corresponding to the search model with the database query condition to form a database query statement, and querying in the database based on the database query statement.
Obtaining a query rule from a search page designer, wherein the query rule comprises a query condition element sub-and a logic operator, traversing the query condition element sub-and analyzing each query operator into an operator algorithm unit, and the method comprises the following steps: identifying a query operator and a binding field; acquiring the data type of the field according to the binding field; deducing a concrete execution algorithm of the operator by using the search model; each operator algorithm unit obtained through analysis is subjected to queue combination according to a logic operator in a query rule, the operator algorithm unit queues are combined with the logic operator, database query conditions are built according to the query rule, and a plurality of conditions are combined, wherein the conditions also comprise nested relations of the conditions; combining the database query conditions with a database table corresponding to the search model to generate a final database query statement, executing query operation in the database by using the generated database query statement, and after the database is connected, executing the database query operation by using the query statement, processing a query result, returning the result to a search executor, and finally presenting a relevant search data result set on a customized query page;
in the embodiment, by acquiring the query rule from the search page designer and analyzing the query condition elements into the operator algorithm unit queue, the system can construct flexible and intelligent query sentences according to user input, so that the system can be more suitable for the diversified query requirements of users instead of fixed predefined query; by combining logic operators in the query rules, the system can support complex query condition construction, including logic relations such as AND, OR AND the like, so that a user can perform more complex AND refined data query operation, AND by combining an operator algorithm unit queue with a database table corresponding to a search model, the system can fully utilize the information of the search model to construct a database query statement in a more intelligent manner, thereby being beneficial to improving the accuracy AND efficiency of query; because of the flexibility and intelligence of the operator algorithm unit queue, the system can better adapt to structural changes of a database table or different query scenes of users, high adaptability is maintained, the users only need to configure query rules in a search page designer, the system can automatically analyze the query rules into database query sentences, the burden of manually writing complex query sentences by the users is reduced, and user experience is improved.
FIG. 2 is a schematic structural diagram of a search page design system based on low code configuration according to an embodiment of the present invention, as shown in FIG. 2, the system includes:
the first unit is used for providing a search page designer and a search model, providing a search column name list for the search page designer based on the search model preset by a table of a database, providing a customized search component by the search page designer, establishing connection with the search model, and creating a search field in the customized search component by dragging and selecting a column name in the search column name list;
a second unit, configured to select at least one query operator from a query operator list adapted to the search field, bind with the search field, and form a query condition element, where in a query rule configurator preset with a logic operator, a rule expression is used to combine the query condition element through the logic operator, so as to form a query rule;
and the third unit is used for providing a search executor comprising an operator algorithm unit, analyzing the query operator in the query condition element into an operator algorithm unit queue based on the query rule acquired from the search page designer, combining a logic operator of the query rule, reconstructing the operator algorithm unit queue into a database query condition, combining a table of a database corresponding to the search model with the database query condition to form a database query statement, and querying in the database based on the database query statement.
In a third aspect of an embodiment of the present invention,
there is provided an electronic device including:
a processor;
a memory for storing processor-executable instructions;
wherein the processor is configured to invoke the instructions stored in the memory to perform the method described previously.
In a fourth aspect of an embodiment of the present invention,
there is provided a computer readable storage medium having stored thereon computer program instructions which, when executed by a processor, implement the method as described above.
The present invention may be a method, apparatus, system, and/or computer program product. The computer program product may include a computer readable storage medium having computer readable program instructions embodied thereon for performing various aspects of the present invention.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and not for limiting the same; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some or all of the technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit of the invention.
Claims (9)
1. The search page design method based on low code collocation is characterized by comprising the following steps:
providing a search page designer and a search model, providing a search column name list for the search page designer based on the search model preset by a table of a database, providing a customized search component by the search page designer, establishing connection with the search model, selecting column names in the search column name list by dragging, and creating a search field in the customized search component;
at least one query operator is selected from a query operator list adapting to the search field and is bound with the search field to form a query condition element, and the query condition element is combined through a logic operator by using a rule expression in a query rule configurator preset with a logic operator to form a query rule;
providing a search executor comprising an operator algorithm unit, analyzing the query operator in the query condition element into an operator algorithm unit queue based on the query rule acquired from the search page designer, combining a logic operator of the query rule, recombining the operator algorithm unit queue into a database query condition, combining a table of a database corresponding to the search model with the database query condition to form a database query statement, and querying in the database based on the database query statement.
2. The method of claim 1, wherein creating a search field in the customized search component by dragging and selecting a column name in the list of search column names, further comprises:
based on the field names collected from the database, a corpus is established, and the corpus is used for training a search intention analyzer;
collecting historical search behavior data of a user in the search page designer, wherein the search intention analyzer learns the user behavior through the historical search behavior data, and the historical search behavior data comprises search preference, a common model, a common field and a use frequency of the user;
setting a search intention text box in the search page designer, inputting a search intention expression language in the search intention text box by a user, analyzing the search intention expression language by the search intention analyzer, extracting keywords and field information, and screening and sequentially presenting user intention fields in combination with the user behaviors.
3. The method of claim 2, wherein training a search intent analyzer using the corpus comprises:
setting a sensing parameter dynamically adjusted according to a context, wherein the sensing parameter is formed by vector connection of an input vector of a target word and an input vector of the whole context through a Concat function, performing linear transformation by using a parameter matrix for adjusting the influence of the context, setting a bias term by considering the mutual influence of the context, and limiting the sensing parameter between 0 and 1 through a sigmoid function, wherein the formula is as follows:
；
wherein,alpha representsThe sensing parameters are used to determine the sensing parameters,W α the parameter matrix is represented by a matrix of parameters,v wt an input vector representing the target word,v context vector representation representing the entire context, concatv wt ,v context ) Representing the target wordwtIs concatenated with the vector representation of the entire context,b α representing the bias term.
4. A method according to claim 3, characterized in that the method further comprises:
the search intention analyzer uses a search language processing model to calculate intention probability by combining the perception parameters, and the formula is as follows:
；
wherein,wcthe context-based words are represented by the terms "a",wtthe target word is represented by a word of interest,P(wc|wt) Representing target wordswtUnder the conditions of appearance, the context wordswcThe probability of the occurrence of the presence of a defect,αrepresenting the perceived parameter(s),v' wc representing context wordswcIs used to determine the output vector of (a),v wt an input vector representing the target word,cnrepresents one of the context words in the whole,v' cn representing context wordscnIs provided.
5. The method of claim 1, wherein selecting at least one query operator from a list of query operators that are adapted to the search field to bind to the search field, further comprises:
acquiring historical binding data of a user through the search list name list and the query operator list, extracting historical binding characteristics, and obtaining a comprehensive recommendation result of the query operator through comprehensive recommendation of the first operator recommendation model and the second operator recommendation model:
according to the historical binding characteristics, using a field name and a field type as input parameters of a first operator recommendation model, randomly extracting samples to form a characteristic subset through the first operator recommendation model, establishing a decision tree, recursively carrying out node splitting to obtain a decision result, integrating a plurality of decision trees, and determining a first recommendation result according to the decision results;
according to the history binding characteristics, taking a target data point formed by a field name and a field type as an input parameter of a second operator recommendation model, calculating the distance between the target data point and other data points through the second operator recommendation model, obtaining a data point list based on the sequential arrangement from small to large, selecting a plurality of data points with the distance smaller than a preset distance threshold value from the data point list as threshold neighbors of the target data point, counting the number of query operators of the threshold neighbors, and taking the query operator with the largest number of query operators as a second recommendation result;
and determining a first recommendation weight corresponding to the first recommendation result and a second recommendation weight corresponding to the second recommendation result by using an information entropy method when comprehensive recommendation is performed, and determining the comprehensive recommendation result by combining the first recommendation result and the second recommendation result.
6. The method of claim 5, wherein determining a first recommendation weight corresponding to the first recommendation and a second recommendation weight corresponding to the second recommendation using an entropy method at the time of integrated recommendation, and wherein determining an integrated recommendation in combination with the first recommendation and the second recommendation comprises:
the weight is calculated by using an information entropy method, and the information entropy is expressed as follows:
；
wherein,Hthe entropy of the information is represented and,x i representing the first of the modelsiThe result of the recommendation is referred to as,P(x i ) Expressed in the model asiProbability of individual recommendation results;
the weight is calculated as follows:
；
wherein,ωthe weight is represented by a weight that,Hthe entropy of the information is represented and,ea base representing natural logarithms;
the formula of the comprehensive recommendation result is as follows:
；
wherein,Tthe result of the integrated recommendation is indicated,ω 1 a first recommendation weight is indicated and a first recommendation weight is indicated,ω 2 a second recommendation weight is indicated as such,R 1 a first recommendation result is indicated and a first recommendation result is indicated,R 2 and representing a second recommendation result.
7. A search page design system based on low code configuration for implementing the search page design method based on low code configuration of any one of the preceding claims 1-6, comprising:
the first unit is used for providing a search page designer and a search model, providing a search column name list for the search page designer based on the search model preset by a table of a database, providing a customized search component by the search page designer, establishing connection with the search model, and creating a search field in the customized search component by dragging and selecting a column name in the search column name list;
a second unit, configured to select at least one query operator from a query operator list adapted to the search field, bind with the search field, and form a query condition element, where in a query rule configurator preset with a logic operator, a rule expression is used to combine the query condition element through the logic operator, so as to form a query rule;
and the third unit is used for providing a search executor comprising an operator algorithm unit, analyzing the query operator in the query condition element into an operator algorithm unit queue based on the query rule acquired from the search page designer, combining a logic operator of the query rule, reconstructing the operator algorithm unit queue into a database query condition, combining a table of a database corresponding to the search model with the database query condition to form a database query statement, and querying in the database based on the database query statement.
8. An electronic device, comprising:
a processor;
a memory for storing processor-executable instructions;
wherein the processor is configured to invoke the instructions stored in the memory to perform the method of any of claims 1 to 6.
9. A computer readable storage medium having stored thereon computer program instructions, which when executed by a processor, implement the method of any of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311850582.8A CN117519702B (en) | 2023-12-29 | 2023-12-29 | Search page design method and system based on low code collocation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311850582.8A CN117519702B (en) | 2023-12-29 | 2023-12-29 | Search page design method and system based on low code collocation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117519702A true CN117519702A (en) | 2024-02-06 |
| CN117519702B CN117519702B (en) | 2024-03-19 |
Family
ID=89766722
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311850582.8A Active CN117519702B (en) | 2023-12-29 | 2023-12-29 | Search page design method and system based on low code collocation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117519702B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109918453A (en) * | 2019-02-13 | 2019-06-21 | 中国三峡建设管理有限公司 | A kind of method and system with Natural Language Search relationship type complex management data of information system |
| CN114168622A (en) * | 2020-09-10 | 2022-03-11 | 北京达佳互联信息技术有限公司 | Data query method and device based on domain specific language |
| CA3138556A1 (en) * | 2020-11-06 | 2022-05-06 | 10353744 Canada Ltd. | Apparatuses, storage medium and method of querying data based on vertical search |
| CN114611010A (en) * | 2022-05-12 | 2022-06-10 | 北京沃丰时代数据科技有限公司 | Commodity search recommendation method and system |
| CN117093708A (en) * | 2023-10-17 | 2023-11-21 | 中电数创(北京)科技有限公司 | Method for intelligently identifying search intention of user and visually displaying search results of element |
-
2023
- 2023-12-29 CN CN202311850582.8A patent/CN117519702B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109918453A (en) * | 2019-02-13 | 2019-06-21 | 中国三峡建设管理有限公司 | A kind of method and system with Natural Language Search relationship type complex management data of information system |
| CN114168622A (en) * | 2020-09-10 | 2022-03-11 | 北京达佳互联信息技术有限公司 | Data query method and device based on domain specific language |
| CA3138556A1 (en) * | 2020-11-06 | 2022-05-06 | 10353744 Canada Ltd. | Apparatuses, storage medium and method of querying data based on vertical search |
| CN114611010A (en) * | 2022-05-12 | 2022-06-10 | 北京沃丰时代数据科技有限公司 | Commodity search recommendation method and system |
| CN117093708A (en) * | 2023-10-17 | 2023-11-21 | 中电数创(北京)科技有限公司 | Method for intelligently identifying search intention of user and visually displaying search results of element |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117519702B (en) | 2024-03-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20210081725A1 (en) | Method, apparatus, server, and user terminal for constructing data processing model | |
| CN101470732B (en) | Auxiliary word stock generation method and apparatus | |
| CN109409533B (en) | Method, device, equipment and storage medium for generating machine learning model | |
| CN105975531B (en) | Robot dialog control method and system based on dialogue knowledge base | |
| CN112199086A (en) | Automatic programming control system, method, device, electronic device and storage medium | |
| CN116719520B (en) | Code generation method and device | |
| CN114547072A (en) | Method, system, equipment and storage medium for converting natural language query into SQL | |
| CN112000330B (en) | Configuration method, device, equipment and computer storage medium of modeling parameters | |
| CN111723192B (en) | Code recommendation method and device | |
| US20140310306A1 (en) | System And Method For Pattern Recognition And User Interaction | |
| CN113297251A (en) | Multi-source data retrieval method, device, equipment and storage medium | |
| CN116049376B (en) | Method, device and system for retrieving and replying information and creating knowledge | |
| CN117520533A (en) | Chart generation method | |
| CN117420998A (en) | Client UI interaction component generation method, device, terminal and medium | |
| CN117519702B (en) | Search page design method and system based on low code collocation | |
| CN116894495A (en) | Method, computer readable medium and system for enhancing machine learning pipeline with annotations | |
| CN111274480B (en) | Feature combination method and device for content recommendation | |
| CN115658885A (en) | Intelligent text labeling method and system, intelligent terminal and storage medium | |
| CN114676155A (en) | Code prompt information determining method, data set determining method and electronic equipment | |
| CN114238581A (en) | Intelligent retrieval system and method based on semantic understanding | |
| Kaur et al. | Towards Visualization Recommendation-A Semi-Automated Domain-Specific Learning Approach. | |
| CN107818134A (en) | A kind of position similarity calculating method, client and server | |
| Lacaba et al. | Medical keyword searching application using word document file as database | |
| CN117390076A (en) | Information processing method and device | |
| CN117910522A (en) | Sentence vector model optimization method, text similarity calculation method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |