List query method and device, electronic equipment and storage medium

Abstract

The invention provides a list query method and device, electronic equipment and a storage medium, and the method comprises the steps: carrying out the feature extraction of a first list through a pre-trained feature extraction model, obtaining a plurality of first features in the first list, and determining the business weight corresponding to each first feature; obtaining a plurality of second features of the plurality of second lists from a preset database; for each second list, determining the feature similarity between the plurality of second features and the first features at the same sorting position in the first list; determining a target similarity between the second list and the first list according to the service weights and the feature similarities of the plurality of first features; and determining a target list of the first list from the plurality of second lists according to the target similarity of the plurality of second lists. According to the embodiment of the invention, lists with similar business requirements can be searched.

Description

Technical Field

[0001] The present application relates to the field of engineering cost technology, and in particular to a list query method, device, electronic device and storage medium. Background Art

[0002] In the related art, the existing cost estimation method is that the cost estimator performs engineering cost estimation based on the features of the list items in the engineering cost list file. In this process, the cost estimator generally applies the quota items of the prepared similar lists and forms a comprehensive unit price. However, since the business requirements of each list are different when it is compiled, the same content has different forms when it is compiled. Therefore, in order to apply the correct quota items, it is necessary to find a compiled list with similar business requirements. In the prior art, it is difficult for cost estimators to find a compiled list with similar business requirements. Summary of the invention

[0003] The main purpose of the embodiments of the present application is to provide a list query method, device, electronic device and storage medium, aiming to find lists with similar business requirements.

[0004] To achieve the above purpose, a first aspect of an embodiment of the present application proposes a list query method, comprising the following steps: Extract features from the first list using a pre-trained feature extraction model to obtain a plurality of first features in the first list, and determine a service weight corresponding to each of the first features; Acquire multiple second features of multiple second lists from a preset database; For each second list, determining a feature similarity between each second feature of the second list and the first feature at the same sorting position in the first list; determining a target similarity between the second list and the first list according to the business weights and the feature similarities of the plurality of first features; A target list of the first list is determined from the plurality of second lists according to the target similarities of the plurality of second lists.

[0005] In one embodiment, the plurality of first features include a plurality of types of the first features; The determining the feature similarity between each of the second features in the second list and the first features at the same sorting position in the first list includes: determining the type to which each of the second features belongs among a plurality of the types; The feature similarity between the first feature and the second feature at the same sorting position in each of the types is determined.

[0006] In one embodiment, the determining the feature similarity between the first feature and the second feature at the same sorting position in each type includes: determining a plurality of target features among the plurality of said first features; Determine the feature similarity between the target feature and the second feature at the same sorting position in each of the types.

[0007] In one embodiment, the multiple types include character type and numeric type; The determining the feature similarity between the first feature and the second feature at the same sorting position in each type includes: For each of the first features in the character type, a semantic similarity analysis is performed based on the first feature and the second feature at the same sorting position in the number type to obtain the feature similarity of the first feature; For each of the first features in the number type, the feature similarity of the first feature is determined according to a difference between the first feature and the second feature at the same sorting position in the character type.

[0008] In one embodiment, performing semantic similarity analysis on the first feature and the second feature at the same sorting position in the character type to obtain the feature similarity of the first feature includes: Performing semantic similarity analysis based on the first feature and the second feature at the same sorting position in the character type to determine the semantic similarity of the first feature; When the semantic similarity is not less than a preset semantic threshold, the preset similarity value is determined as the feature similarity.

[0009] In one embodiment, determining the feature similarity of the first feature according to the difference between the first feature and the second feature at the same sorting position in the data type includes: Determine a feature having the largest value between the first feature and the second feature at the same sorting position in the data type; The feature similarity of the first feature is determined according to the feature with the largest value and the difference.

[0010] In one embodiment, determining the service weight corresponding to each of the first features includes: Obtaining service usage information of the first list; Performing a list influence analysis on each of the first features according to the service usage information to obtain a list influence of each of the first features on the first list; The business weight of each of the first features is determined according to the list influence of each of the first features.

[0011] To achieve the above-mentioned purpose, a second aspect of an embodiment of the present application provides a list query device, including: A first feature extraction module, configured to extract features from the first list by using a pre-trained feature extraction model, obtain a plurality of first features in the first list, and determine a business weight corresponding to each of the first features; A second feature acquisition module, used to acquire multiple second features of multiple second lists from a preset database; a target similarity determination module, configured to determine, for each second list, feature similarities between a plurality of the second features and the first features at the same sorting position in the first list; and determine the target similarity between the second list and the first list according to the business weights and feature similarities of the plurality of the first features; A target list determination module is used to determine a target list of the first list from a plurality of the second lists according to the target similarities of the plurality of the second lists.

[0012] To achieve the above objectives, a third aspect of an embodiment of the present application proposes an electronic device, which includes a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the method described in the first aspect is implemented.

[0013] To achieve the above-mentioned purpose, the fourth aspect of an embodiment of the present application proposes a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method described in the first aspect is implemented.

[0014] The present application proposes a list query method, device, electronic device and storage medium, the method comprising: extracting features from a first list through a pre-trained feature extraction model to obtain multiple first features in the first list, and determining the business weight corresponding to each of the first features; obtaining multiple second features of multiple second lists from a preset database; for each of the second lists, determining the feature similarity between each of the second features of the second list and the first feature at the same sorting position in the first list; determining the target similarity between the second list and the first list based on the business weights and feature similarities of the multiple first features; and determining the target list of the first list from the multiple second lists based on the target similarities of the multiple second lists. By determining the business weight of each first feature, the influence representation of each first feature in satisfying the business needs of the first list can be determined; by determining the feature similarity between multiple second features and the first feature at the same sorting position in the first list, the target similarity between the second list and the first list is determined according to the business weights and feature similarities of the multiple first features. In this way, in the process of determining the target similarity, the influence of each first feature on the target similarity and the influence representation of each first feature in satisfying the business needs on the target similarity can be taken into account, so that the business needs can be integrated into the process of finding similar lists, and the target similarity can carry business need information, so that lists with similar business needs can be found through target similarity. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 is a flow chart of a list query method provided by an embodiment of the present application; Figure 2 yes Figure 1 A flowchart of an embodiment of a sub-step of step 130; Figure 3 yes Figure 2 A flowchart of an embodiment of a sub-step of step 220; Figure 4 yes Figure 2 A flowchart of another sub-step embodiment of step 220; Figure 5 yes Figure 4 A flowchart of an embodiment of a sub-step of step 410; Figure 6 yes Figure 4 A flowchart of an embodiment of a sub-step of step 420; Figure 7 yes Figure 1 A flowchart of an embodiment of a sub-step of step 110; Figure 8is a schematic diagram of the structure of the inventory query device provided in an embodiment of the present application; Fig. 9 It is a schematic diagram of the hardware structure of the electronic device provided in the embodiment of the present application. DETAILED DESCRIPTION

[0016] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0017] It should be noted that, although the functional modules are divided in the device schematic diagram and the logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the device or the order in the flowchart. The terms "first", "second", etc. in the specification, claims and the above drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence.

[0018] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.

[0019] In the related art, the existing cost estimation method is that the cost estimator performs engineering cost estimation based on the features of the list items in the engineering cost list file. In this process, the cost estimator generally applies the quota items of the prepared similar lists and forms a comprehensive unit price. However, since the business requirements of each list are different when it is compiled, the same content has different forms when it is compiled. Therefore, in order to apply the correct quota items, it is necessary to find a compiled list with similar business requirements. In the prior art, it is difficult for cost estimators to find a compiled list with similar business requirements.

[0020] In order to be able to find lists with similar business needs, an embodiment of the present application provides a list query method, a list query device, an electronic device and a computer-readable storage medium, the method comprising: extracting features from a first list through a pre-trained feature extraction model to obtain multiple first features in the first list, and determining the business weight corresponding to each first feature; obtaining multiple second features of multiple second lists from a preset database; for each second list, determining the feature similarity between each second feature of the second list and the first feature at the same sorting position in the first list; determining the target similarity between the second list and the first list based on the business weights and feature similarities of the multiple first features; and determining the target list of the first list from the multiple second lists based on the target similarities of the multiple second lists. By determining the business weight of each first feature, the influence representation of each first feature in satisfying the business needs of the first list can be determined; by determining the feature similarity between multiple second features and the first feature at the same sorting position in the first list, the target similarity between the second list and the first list is determined according to the business weights and feature similarities of the multiple first features. In this way, in the process of determining the target similarity, the influence of each first feature on the target similarity and the influence representation of each first feature in satisfying the business needs on the target similarity can be taken into account, so that the business needs can be integrated into the process of finding similar lists, and the target similarity can carry business need information, so that lists with similar business needs can be found through target similarity.

[0021] An inventory query method, device, electronic device and storage medium provided in the embodiments of the present application are specifically described through the following embodiments. First, the inventory query method in the embodiments of the present application is described.

[0022] The inventory query method provided in the embodiment of the present application can be applied to the terminal, can also be applied to the server side, and can also be software running in the terminal or the server side. In some embodiments, the terminal can be a smart phone, a tablet computer, a laptop computer, a desktop computer, etc.; the server side can be configured as an independent physical server, or a server cluster or distributed system composed of multiple physical servers, and can also be configured as a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms; the software can be an application that implements the inventory query method, etc., but is not limited to the above forms.

[0023] The present application can be used in many general or special computer system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, etc. The present application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present application can also be practiced in distributed computing environments, in which tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media including storage devices.

[0024] See also Figure 1 , Figure 1 The flowchart of an inventory query method provided by an embodiment of the present application is shown. In the embodiment of the present application, the inventory query method may include steps 110 to 140.

[0025] Step 110: extracting features from the first list using a pre-trained feature extraction model to obtain a plurality of first features in the first list, and determining a service weight corresponding to each first feature; Step 120: Acquire multiple second features of multiple second lists from a preset database; Step 130: for each second list, determining the feature similarity between each second feature of the second list and the first feature at the same sorting position in the first list; determining the target similarity between the second list and the first list based on the business weights and feature similarities of the plurality of first features; Step 140: Determine a target list of the first list from the plurality of second lists according to the target similarities of the plurality of second lists.

[0026] In one embodiment, the first list refers to a form that is in a compilation state and is used to record the cost content of a sub-project in an overall engineering project. The first list may include a project code, a project name, and a project feature (i.e., the first feature mentioned in step 110), wherein the project code refers to a specific code used to indicate the project corresponding to the first list in the engineering cost standard specification, the project name refers to the name used to indicate the project corresponding to the first list, and the project feature refers to a part of the actual engineering content (such as transportation distance, construction materials, construction methods, etc.) used to describe the project corresponding to the first list. When it is recorded in the first list, it may or may not contain indicative words, which are not specifically limited here. In addition, the order between multiple first features complies with the compilation specification, for example, "soil category: Class I soil" is located before "abandoned soil transportation distance: within 2m", and "abandoned soil transportation distance: within 2m" is not located before "soil category: Class I soil".

[0027] For example, the first list can be in the following format (from left to right are project code, project name and project characteristics, and each type of content is represented by ","): "010101001, leveling site, soil type: Class I soil, distance for waste soil transportation: ≤2m", among which "soil type" is an indicative term for the characteristic of "Class I soil", and "distance for waste soil transportation" is an indicative term for "≤2m".

[0028] For another example, the first list may be in the following format: "010101001, level site, type 1 soil, ≤2m", where the first feature does not contain any indicative words.

[0029] In one embodiment, the feature extraction model refers to a model used to extract features of all items included in the list. The feature extraction model may be a large language model (LLM), a model trained based on a Transformer model architecture, etc., which is not specifically limited here.

[0030] In one embodiment, the feature extraction model can be a model formed by combining a bidirectional long-short term memory network (Bidirectional Long-Short Term Memory, Bi-LSTM) and a conditional random field (Conditional Random Field, CRF). The feature extraction model can be set using an input layer, a coding layer, and a model architecture that has been tested. The initial value of the learning rate of the feature extraction model can be set to 0.001, the batch size can be initially set to 32, the optimizer (Optimizer) selected during the training process can be an adaptive moment estimation (Adaptive Moment Estimation, Adam) algorithm, and Dropout with an initial parameter setting of 0.25 can be selected to prevent overfitting of the feature extraction model. The feature extraction model is trained using a training set, and after the training is completed, the parameters are adjusted using a validation set, and then evaluated using a test set. When the precision (Precision), recall (Recall) and F1 score of the model are all greater than 85%, the final feature extraction model is obtained, wherein the ratio between the training set, the validation set, and the test set can be 80%:10%:10%.

[0031] In one embodiment, the business weight refers to the influence coefficient of the project characteristics in a list on the construction content of the sub-project represented by the list in the business needs it needs to meet. For example, for the project of leveling the site, when the business needs are to focus on the distance limit, the distance of waste soil transportation is relatively the most important, and its business weight can be set to 0.5. Among them, the business weight can be determined by the knowledge map related to the project cost, or it can be pre-set, and it is not specifically limited here.

[0032] In one embodiment, the second list refers to a form that has been compiled and is used to record the cost content of a sub-project in an overall engineering project. Among them, like the first list, the second list can also include project codes, project names and project characteristics (that is, the second characteristics mentioned in step 120). Among them, since the second list queried is generally a second list for the same construction content as the first list, the second list with different construction contents has no reference significance (for example, the quota application of leveling the site should refer to the quota of other leveling site lists, and cannot refer to the quota of frozen soil excavation lists). Therefore, the project code and project name of each second list obtained from the preset database are the same as the project code and project name of the first list. In addition, the order between multiple second characteristics also conforms to the compilation specifications mentioned above.

[0033] In one embodiment, obtaining multiple second features of multiple second lists refers to an operation of obtaining multiple second features included in each second list in the multiple second lists. The second features can be stored in a preset database as a separate type of information. In the process of executing step 120, all the second features of a specific second list can be directly called from the preset database without using a feature extraction model to extract the second list separately. In addition, in the process of storing the second list in the second list into the preset database, the feature extraction model can be used first for rapid extraction, thereby improving the efficiency of sorting the list.

[0034] In one embodiment, determining the feature similarity between each second feature of the second list and the first feature at the same sorting position in the first list refers to first determining the sorting position of each second feature among the multiple second features in the second list, then determining the first feature at the same position among the multiple first features in the first list based on the sorting position of each second feature among the multiple second features, and then determining the feature similarity between each second feature and the corresponding first feature. The sorting position here can be a sorting position classified by type or a sorting position not classified by type, in which the type can be a feature attribute, a data type, etc., which is not specifically limited here. In addition, the sorting position corresponding to each feature can be represented by a position identifier.

[0035] For example, assuming that there is a second list, whose content is "010101001, leveling site, soil category: Class I soil, distance for waste soil transportation: within 2m", then for this second list, "soil category: Class I soil" ranks first, and "distance for waste soil transportation: within 2m" ranks second. Suppose that there is a first list, whose content is "010101001, leveling site, soil category: Class III soil, distance for waste soil transportation: 3km", then for this first list, "soil category: Class III soil" ranks first, and "distance for waste soil transportation: 3km" ranks second. Then, in the process of executing step 130, it is necessary to determine the feature similarity between "soil category: Class I soil" and "soil category: Class III soil", as well as to determine the feature similarity between "waste soil transportation distance: within 2m" and "waste soil transportation distance: 3km".

[0036] For another example, suppose there is a second list, and its content is "010101001, leveling site, soil category: Class 1 soil", then for this second list, "soil category: Class 1 soil" ranks first. Suppose there is a first list, and its content is "010101001, leveling site, soil category: Class 3 soil, AAA leveling method", then for this first list, "soil category: Class 3 soil" ranks first and "AAA leveling method" ranks second. Then, in the process of executing step 130, it is necessary to determine the feature similarity between "soil category: Class 1 soil" and "soil category: Class 3 soil". Since the second list does not have a second feature in the second ranking, the similarity corresponding to "AAA leveling method" is 0.

[0037] In one embodiment, determining the target similarity between the second list and the first list based on the business weights and feature similarities of multiple first features refers to first determining the similarity corresponding to each first feature based on the feature similarity and business weight of each first feature, and then determining the target similarity between the second list and the first list based on the similarity corresponding to each first feature. Determining the target similarity between the second list and the first list based on the similarity corresponding to each first feature can be achieved by taking a specific value therein, superimposing each similarity, etc., which is not specifically limited here.

[0038] In one embodiment, determining the target list of the first list from the multiple second lists according to the target similarities of the multiple second lists refers to sorting the target lists according to the target similarity values ​​of each second list and determining the target list from the multiple second lists according to the sorting. The target list may be one or more, which is not specifically limited here.

[0039] See also Figure 2 , Figure 2 Shows Figure 1 In the process of a sub-step embodiment of step 130 in the embodiment, in one embodiment, the plurality of first features include a plurality of types of first features. Step 130 may include the following steps.

[0040] Step 210: Determine the type to which each second feature belongs among the multiple types; Step 220: Determine the feature similarity between the first feature and the second feature at the same sorting position in each type.

[0041] In one embodiment, a preset database records type information corresponding to each second feature. In the process of determining the type to which each second feature belongs among multiple types, the type information corresponding to each second feature can be retrieved from the preset database according to each second feature, and then the type to which each second feature belongs among multiple types can be determined according to the type information corresponding to each second feature.

[0042] In one embodiment, determining the feature similarity between the first feature and the second feature at the same ranking position in each type refers to an operation of determining, for each type of first feature and second feature, the feature similarity between each first feature and the second feature at the same ranking position in the type. The same ranking position in each type refers to the ranking position of the feature in the type after the feature is classified into the type.

[0043] For example, suppose the content of a first list is "aaa project number, bbb project name, CA1, CA2, CB1", and the content of a second list is "aaa project number, bbb project name, CA3, CA4, CB2", where A and B are type designations, then the features of type A in the first list are CA1 and CA2, and the feature of type B is CB1; the features of type A in the second list are CA3 and CA4, and the feature of type B is CB2, then, when determining the feature similarity, the feature similarities between CA1 and CA3, CA2 and CA4, and CB1 and CB2 are determined respectively.

[0044] For another example, suppose the content of a first list is "aaa project number, bbb project name, cA1, cB1, cC1", and the content of a second list is "aaa project number, bbb project name, cA2, cB2", where A, B, and C are type references, then the features of type A in the first list include cA1, the features of type B include cB1, and the features of type C include cC1; the features of type A in the second list include cA2, the features of type B include cB2, and there are no features of type C, then, when determining the feature similarity, the feature similarities between cA1 and cA2, cB1 and cB3 are determined respectively, and the feature similarity of cC1 is determined to be 0.

[0045] Due to compilation habits, some second lists may have the number of second features not corresponding to the number of first features, resulting in position alignment differences between the second features and the first features, which may lead to the first feature and the second feature that need to determine the similarity not corresponding (for example, at the same position, one is material and the other is distance), resulting in subsequent feature similarity calculation errors. Therefore, by calculating the feature similarity between the first feature and the second feature at the same sorting position in each type, the correspondence between the first feature and the second feature can be improved by type, thereby improving the accuracy of the feature similarity between the first feature and the second feature.

[0046] See also Figure 3 , Figure 3 Shows Figure 2In the process of a sub-step embodiment of step 220, in one embodiment, step 220 may include the following steps.

[0047] Step 310: determining a plurality of target features from a plurality of first features; Step 320: Determine the feature similarity between the target feature and the second feature at the same sorting position in each type.

[0048] In one embodiment, the target feature refers to the first feature that can substantially represent the specific content of the first list. Specifically, when compiling the list, due to the different compilation habits of cost personnel, some explanatory first features may be added. These first features have no substantial impact on the first list and are not the target features mentioned in this application. On the contrary, some first features that can clarify the specific content of the first list are the target features mentioned in this application.

[0049] In one embodiment, determining the feature similarity between the target feature and the second feature at the same sorting position in each type refers to an operation of determining, for each type of target feature and second feature, the feature similarity between each target feature and the second feature at the same sorting position in the type.

[0050] For example, suppose the content of a first list is "aaa project number, bbb project name, CA1, CB1, Q1", and the content of a second list is "aaa project number, bbb project name, CA2, CB2", where A and B are type references, then the features of type A in the first list are CA1, and the features of type B are CB1, and CA1 and CB1 are both target features; the features of type A in the second list are CA2, and the features of type B are CB2, then, when determining the feature similarity, the feature similarity between CA1 and CA2, and between CB1 and CB2 are determined separately.

[0051] The feature similarity calculation process is performed through the target feature, so that the accuracy of the target similarity can be improved through the key elements in the first list, thereby reducing or even eliminating the influence of non-target features on the target similarity. In addition, since the feature similarity calculation process is performed for the target feature, the calculation efficiency of the target similarity can be improved, so that the target similarity can be determined more quickly.

[0052] See also Figure 4 , Figure 4 Shows Figure 2 In another embodiment of step 220, in one embodiment, the multiple types include character type and number type. Step 220 may include the following steps.

[0053] Step 410: for each first feature in the character type, a semantic similarity analysis is performed based on the first feature and the second feature at the same sorting position in the character type to obtain feature similarity of the first feature; Step 420: For each first feature in the digital type, determine the feature similarity of the first feature according to the difference between the first feature and the second feature in the same sorting position in the character type.

[0054] In one embodiment, the feature of the character type refers to the word feature that the described content is composed of characters, such as the "first type of soil" mentioned above; the first feature of the data type refers to the data feature that the described content is composed of numbers, such as the "3km" mentioned above.

[0055] In one embodiment, for each first feature in a character type, a semantic similarity analysis is performed based on the first feature and the second feature at the same sorting position in the character type to obtain the feature similarity of the first feature, which means first determining the second feature at the same sorting position of each first feature in the character type among multiple second features in the character type, and then performing a semantic similarity analysis on the first feature of each character type and the second feature of the corresponding character type to obtain the feature similarity of the first feature. The semantic similarity analysis may be to determine the similarity using semantic features, or to determine the similarity by determining whether the features are strictly identical in text, etc., which is not specifically limited here.

[0056] In one embodiment, for each first feature in the digital type, the feature similarity of the first feature is determined based on the difference between the first feature and the second feature in the same sorting position in the character type, which means first determining the second feature in the same sorting position of each first feature in the digital type among multiple second features in the character type, then performing a difference operation on the first feature of each digital type and the second feature of the corresponding digital type to obtain the numerical difference between the first feature of each digital type and the second feature of the corresponding digital type, and then determining the feature similarity corresponding to the first feature of each digital type based on the numerical difference corresponding to the first feature of each digital type. There are various ways to determine the feature similarity based on the numerical difference, which can be queried through a preset table, or the similarity can be obtained by performing a certain numerical conversion on the numerical difference, etc., which are not specifically limited here.

[0057] In one embodiment, in the process of determining the feature similarity of the first feature based on the difference between the first feature and the second feature at the same sorting position in the digital type, the numerical units of the first feature and the second feature at the same sorting position in the digital type can be determined first. When the numerical units of the two are different, the similarity is determined to be 0. When the numerical units of the two are the same, the difference between the two is calculated. By determining whether the difference is calculated by determining whether the numerical units are the same, the occurrence of a high similarity due to different units and similar numerical values ​​can be reduced, thereby improving the accuracy of the similarity of digital type features.

[0058] By calculating the feature similarity for character type features and number type features respectively, the similarity between different types of features can be determined using different similarity determination methods based on the characteristics of the different types of features. The similarity determination method used for each type of feature matches the characteristics of the feature of that type, thereby making the feature similarity of each feature of each different type more accurate.

[0059] See also Figure 5 , Figure 5 Shows Figure 4 In the process of a sub-step embodiment of step 410, in one embodiment, step 410 may include the following steps.

[0060] Step 510: performing semantic similarity analysis based on the first feature and the second feature at the same sorting position in the character type to determine the semantic similarity of the first feature; Step 520: When the semantic similarity is not less than a preset semantic threshold, the preset similarity value is determined as the feature similarity.

[0061] In one embodiment, in the process of performing semantic similarity analysis based on the first feature and the second feature at the same sorting position in the character type to determine the semantic similarity of the first feature, the semantic feature vectors of the first feature and the second feature can be extracted respectively to obtain the semantic feature vectors of the first feature and the second feature, and then the semantic similarity of the first feature is determined based on the semantic feature vectors of the first feature and the second feature. The semantic similarity can be determined by the distance between the two semantic feature vectors in the feature space.

[0062] In one embodiment, the semantic threshold refers to a threshold used to determine whether the first feature and the second feature have equivalent semantics. The specific value of the semantic threshold can be various, such as 0.9, 0.95, etc., which is not specifically limited here.

[0063] In one embodiment, after determining the semantic similarity of the first feature, the step may further include comparing the semantic similarity with a preset semantic threshold, and determining whether the first feature is the same as the second feature based on the magnitude relationship between the semantic similarity and the preset semantic threshold. When the semantic similarity is not less than the preset semantic threshold, the first feature is the same as the second feature; when the semantic similarity is less than the preset semantic threshold, the first feature is different from the second feature.

[0064] In one embodiment, when the first feature is different from the second feature, the feature similarity is determined to be 0.

[0065] In one embodiment, the process of determining whether the first feature is identical to the corresponding second feature to determine the feature similarity between the first feature and the second feature can be represented by formula (1).

[0066] (1); in, It refers to the feature similarity between the first feature and the second feature at the i-th position in the character type, where i is a positive integer; refers to the first feature at the i-th position among the first features of the plurality of character types in the first list; It refers to the second feature at the i-th position among the second features of the plurality of character types in the second list.

[0067] Whether the first feature and the second feature are identical is determined by the semantic similarity between the two features and a preset semantic threshold, and the feature similarity between the two is determined based on this result, so that the semantic similarity between the first feature and the second feature does not need to be directly used as the feature similarity, so that the same first feature and the second feature can be limited to a preset similarity value, thus avoiding the influence of too low or too high semantic similarity on the subsequent calculation of the similarity between the first list and the second list, so that the similarity between the first list and the second list can be made more accurate.

[0068] In one embodiment, step 410 may specifically determine whether the first feature and the second feature at the same sorting position in the character type are strictly identical in text, and based on formula (1), determine the feature similarities corresponding to the same and different features.

[0069] See also Figure 6 , Figure 6 Shows Figure 4 In the process of a sub-step embodiment of step 420, in one embodiment, step 420 may include the following steps.

[0070] Step 610: Determine the feature with the largest value between the first feature and the second feature at the same sorting position in the data type; Step 620: Determine the feature similarity of the first feature according to the feature with the largest value and the difference.

[0071] In one embodiment, in the process of determining the feature similarity of the first feature according to the feature with the largest value and the difference, the coefficient of the difference can be first determined according to the value of the feature, and then determined based on the coefficient and the difference.

[0072] In one embodiment, the process of determining the feature similarity of the first feature according to the feature with the largest value and the difference can be expressed by formula (2).

[0073] (2); in, It refers to the feature similarity between the first feature and the second feature at the jth position in the digital type, where j is a positive integer; refers to the first feature at the jth position among the first features of the plurality of digital types in the first list; refers to the second feature at the j-th position among the second features of the plurality of digital types in the second list; It refers to the feature with the largest value between the first feature and the second feature at the jth position.

[0074] The feature similarity of the first feature is determined by the feature with the largest value in the first feature and the second feature of the digital type and the difference between the two. In this way, the situation where the semantic similarity analysis method cannot extract the semantics of the first feature and the second feature of the digital type and thus cannot determine the feature similarity between the first feature and the second feature of the digital type can be avoided. Therefore, the feature similarity between the first feature and the second feature of the digital type can be determined more accurately. In this way, in the process of determining the target similarity between the first list and the second list, the influence of the feature similarity between the first feature and the second feature of the digital type on the target similarity can be taken into account, which helps to improve the accuracy of the target similarity.

[0075] In one embodiment, based on multiple types including character type and number type, according to the business weights and feature similarities of multiple first features, the process of determining the target similarity between the second list and the first list can be represented by formula (3).

[0076] (3); in, It refers to the target similarity between the second list and the first list; It refers to the business weight corresponding to the first feature at the i-th position in the character type; It refers to the feature similarity of the first feature at the i-th position in the character type; It refers to the business weight corresponding to the first feature at the jth position in the digital type; It refers to the feature similarity of the first feature at the jth position in the digital type; refers to the number of first features of the character type; Refers to the number of the first feature of numeric type.

[0077] The feature similarities of each first feature are superimposed by the business weights of each first feature and then averaged. In this way, in the process of determining the target similarity, the influence of the similarity between each first feature and its corresponding second feature on the target similarity can be taken into account, so that the target similarity can more accurately reflect the similarity between the first list and the second list, which helps to improve the accuracy of the similarity analysis process between the second list and the first list.

[0078] See also Figure 7 , Figure 7 Shows Figure 1 In one embodiment, step 110 may include the following steps.

[0079] Step 710: Obtaining service usage information of the first list; Step 720: Perform inventory influence analysis on each first feature according to the service usage information to obtain the inventory influence of each first feature on the first inventory; Step 730: Determine the business weight of each first feature according to the list influence of each first feature.

[0080] In one embodiment, the business usage information refers to information used to indicate the business requirements for using the first list. The business usage information may indicate that the first list is used during the contract preparation stage of the project, and may indicate that the material transportation distance should be paid attention to during the preparation process, etc., which is not specifically limited here.

[0081] In one embodiment, the list influence refers to the degree of influence of the first feature in the process when the first list meets the business usage requirements indicated by the business usage information. For example, when the business usage information indicates that attention should be paid to the material transportation distance during the compilation process, the first feature indicating the distance has a greater influence on the first list than other first features, that is, the first feature indicating the distance has a greater list influence than other first features.

[0082] In one embodiment, during the inventory influence analysis, the inventory influence of each first feature on the first inventory can be obtained by inputting the business usage information and each first feature into a pre-trained decision model, or by pre-constructing an inventory influence table that can record different features under the business usage business requirements for different usage business requirements, and then determining a specific inventory influence table through the business usage information, and then determining the inventory influence of each first feature on the first inventory through the specific inventory influence table, and so on, which is not specifically limited here.

[0083] In one embodiment, in the process of determining the service weight of each first feature according to the list influence of each first feature, the list influence of each first feature can be normalized to obtain the normalized value of the list influence of each first feature, and then the normalized value of the list influence of each first feature is used as the service weight of each first feature. The normalization process can be implemented by algorithms such as sigmoid and softmax, which are not specifically limited here.

[0084] In one embodiment, in the process of determining the service weight of each first feature according to the list influence of each first feature, the list influence of each first feature can be directly used as the service weight of each first feature.

[0085] The list influence of the first feature is determined through business usage information, and then the business weight is determined through the list influence. In this way, the business weight can be adaptively changed following the business needs of the first list, so that the target similarity can take into account the influence of the business needs of the first list, thereby making the target similarity closer to the business needs of the first list, thereby improving the generalization ability of the list query.

[0086] See also Figure 8 The embodiment of the present application further provides a list query device, which can implement the list query method of the first aspect above. The list query device 800 includes: The first feature extraction module 810 may be used to extract features from the first list using a pre-trained feature extraction model, obtain multiple first features in the first list, and determine a business weight corresponding to each first feature; The second feature acquisition module 820 may be used to acquire multiple second features of multiple second lists from a preset database; The target similarity determination module 830 may be used to determine, for each second list, the feature similarity of each second feature of the second list with the first feature at the same sorting position in the first list; determine the target similarity between the second list and the first list based on the business weights and feature similarities of the plurality of first features; The target list determination module 840 may be configured to determine a target list of the first list from the plurality of second lists according to target similarities of the plurality of second lists.

[0087] The specific implementation of the inventory query device is basically the same as the specific implementation of the above-mentioned inventory query method, and will not be repeated here.

[0088] The embodiment of the present application also provides an electronic device, which may include a memory and a processor, wherein the memory stores a computer program, and the processor implements the above-mentioned inventory query method when executing the computer program. The electronic device may be any intelligent terminal including a tablet computer, a vehicle-mounted computer, etc.

[0089] See also Fig. 9 , Fig. 9 The hardware structure of an electronic device of another embodiment is illustrated. The electronic device 900 includes: The processor 901 may be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (Application Specific Integrated Circuit, ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of the present application; The memory 902 can be implemented in the form of a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 902 can store an operating system and other application programs. When the technical solution provided in the embodiment of this specification is implemented by software or firmware, the relevant program code is stored in the memory 902, and the processor 901 calls and executes the inventory query method of the embodiment of this application; Input / output interface 903, used to implement information input and output; Communication interface 904, used to realize communication interaction between the device and other devices, which can be realized through wired mode (such as USB, network cable, etc.) or wireless mode (such as mobile network, WIFI, Bluetooth, etc.); A bus 905 that transmits information between various components of the device (e.g., the processor 901, the memory 902, the input / output interface 903, and the communication interface 904); The processor 901 , the memory 902 , the input / output interface 903 and the communication interface 904 are connected to each other in communication within the device via a bus 905 .

[0090] An embodiment of the present application further provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the above-mentioned inventory query method is implemented.

[0091] The memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer executable programs. In addition, the memory may include a high-speed random access memory, and may also include a non-transitory memory, such as at least one disk storage device, a flash memory device, or other non-transitory solid-state storage device. In some embodiments, the memory may include a memory remotely disposed relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0092] The embodiments described in the embodiments of the present application are intended to more clearly illustrate the technical solutions of the embodiments of the present application, and do not constitute a limitation on the technical solutions provided in the embodiments of the present application. Those skilled in the art will appreciate that with the evolution of technology and the emergence of new application scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.

[0093] Those skilled in the art will appreciate that the technical solutions shown in the figures do not constitute a limitation on the embodiments of the present application, and may include more or fewer steps than shown in the figures, or a combination of certain steps, or different steps.

[0094] The device embodiments described above are merely illustrative, and the units described as separate components may or may not be physically separated, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0095] Those skilled in the art will appreciate that all or some of the steps in the methods disclosed above, and the functional modules / units in the systems and devices may be implemented as software, firmware, hardware, or a suitable combination thereof.

[0096] The terms "first", "second", "third", "fourth", etc. (if any) in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0097] It should be understood that in the present application, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the objects associated before and after are in an "or" relationship. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0098] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the above units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0099] The units described above as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0100] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0101] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including multiple instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of various embodiments of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, referred to as ROM), random access memory (Random Access Memory, referred to as RAM), disk or optical disk and other media that can store programs.

[0102] The preferred embodiments of the present invention are described above with reference to the accompanying drawings, but the scope of the rights of the present invention is not limited thereto. Any modification, equivalent substitution and improvement made by a person skilled in the art without departing from the scope and essence of the present invention should be within the scope of the rights of the present invention.

Claims

1. A method for querying a list, characterized in that: The following steps are involved: Extract features from the first list using a pre-trained feature extraction model to obtain a plurality of first features in the first list, and determine a service weight corresponding to each of the first features; Acquire multiple second features of multiple second lists from a preset database; For each second list, determining a feature similarity between each second feature of the second list and the first feature at the same sorting position in the first list; determining a target similarity between the second list and the first list according to the business weights and the feature similarities of the plurality of first features; A target list of the first list is determined from the plurality of second lists according to the target similarities of the plurality of second lists.

2. The method according to claim 1, characterized in that The plurality of first features include a plurality of types of the first features; The determining the feature similarity between each of the second features in the second list and the first features at the same sorting position in the first list includes: determining the type to which each of the second features belongs among a plurality of the types; The feature similarity between the first feature and the second feature at the same sorting position in each of the types is determined.

3. The method according to claim 2, characterized in that The determining the feature similarity between the first feature and the second feature at the same sorting position in each type includes: determining a plurality of target features among the plurality of said first features; Determine the feature similarity between the target feature and the second feature at the same sorting position in each of the types.

4. The method according to claim 2, characterized in that: Multiple types include character types and numeric types; The determining the feature similarity between the first feature and the second feature at the same sorting position in each type includes: For each of the first features in the character type, a semantic similarity analysis is performed based on the first feature and the second feature at the same sorting position in the character type to obtain the feature similarity of the first feature; For each of the first features in the digital type, the feature similarity of the first feature is determined according to a difference between the first feature and the second feature at the same sorting position in the digital type.

5. The method according to claim 4, characterized in that The performing semantic similarity analysis on the first feature and the second feature at the same sorting position in the character type to obtain the feature similarity of the first feature includes: Performing semantic similarity analysis based on the first feature and the second feature at the same sorting position in the character type to determine the semantic similarity of the first feature; When the semantic similarity is not less than a preset semantic threshold, the preset similarity value is determined as the feature similarity.

6. The method according to claim 5, characterized in that The determining the feature similarity of the first feature according to the difference between the first feature and the second feature at the same sorting position in the data type includes: Determine a feature having the largest value between the first feature and the second feature at the same sorting position in the data type; The feature similarity of the first feature is determined according to the feature with the largest value and the difference.

7. The method according to claim 1, characterized in that The determining the service weight corresponding to each of the first features includes: Obtaining service usage information of the first list; Performing a list influence analysis on each of the first features according to the service usage information to obtain a list influence of each of the first features on the first list; The business weight of each of the first features is determined according to the list influence of each of the first features.

8. A list query device, characterized in that: include: A first feature extraction module, configured to extract features from the first list by using a pre-trained feature extraction model, obtain a plurality of first features in the first list, and determine a business weight corresponding to each of the first features; A second feature acquisition module, used to acquire multiple second features of multiple second lists from a preset database; a target similarity determination module, configured to determine, for each second list, a feature similarity between each second feature of the second list and the first feature at the same sorting position in the first list; and determine a target similarity between the second list and the first list according to the business weights and the feature similarities of a plurality of the first features; A target list determination module is used to determine a target list of the first list from a plurality of the second lists according to the target similarities of the plurality of the second lists.

9. An electronic device, characterized in that: The electronic device comprises a memory and a processor, the memory stores a computer program, and the processor implements the inventory query method according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the inventory query method according to any one of claims 1 to 7 is implemented.

Images