Method and device for data attribution, electronic equipment and program product

Abstract

The embodiment of the invention relates to a data attribution method and device, electronic equipment and a program product. The method comprises the steps that data in a text description form is obtained, an analysis tree is generated based on the data in the text description form and a preset dimension, and the analysis tree comprises at least one of node attribution data and whole tree attribution data. The method further comprises the step of cutting the analysis tree based on the first parameter and the second parameter to obtain cut data in a multi-way tree structure form. The method further includes normalizing and serializing the cut data in the form of the multi-way tree structure to generate normalized and serialized data, and presenting the normalized and serialized data in the form of text description. Further, the method includes generating, by a language model, an attribution summary for the data based on the normalized serialized data. Therefore, according to the scheme of the embodiment of the invention, attribution aiming at the data can be automatically performed for the user in mass data, and the attribution is presented in a concise character, so that the user experience is improved.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and more particularly to a method, device, electronic device, and computer program product for data attribution. Background Art

[0002] With the advent of the big data era, the amount of data is growing rapidly and becoming increasingly complex, which makes data analysis more and more important in all walks of life. Through efficient processing and analysis of massive data, data analysis can help users deeply explore the value of data, optimize decision-making processes, and improve operational efficiency. Therefore, the importance of data analysis is constantly increasing.

[0003] Given the explosive growth of data volume, how to extract valuable information from such a large and complex data system has become a key challenge in data analysis. Data attribution, as a crucial part of the data analysis process, has become increasingly important in this context. Simply put, data attribution is the process of determining the causes and influencing factors behind data results. With accurate data attribution, we can have a deeper understanding of the logic behind the data and provide a solid and reliable basis for decision-making. Summary of the invention

[0004] Embodiments of the present disclosure provide a method, apparatus, electronic device, and computer program product for data attribution.

[0005] According to a first aspect of the present disclosure, a method for data attribution is provided. The method includes obtaining data in the form of text description, wherein the data includes changes in the data within a date range. The method includes generating a parsing tree based on the data in the form of text description and a preset dimension, wherein the parsing tree includes at least one of node attribution data and whole tree attribution data, the node attribution data indicates that the data is presented in the form of node hierarchical nesting, each node in the node attribution data includes a key numerical description of the node, and the whole tree attribution data is determined based on the node's disassembly path. The method also includes pruning the parsing tree based on a first parameter and a second parameter to obtain pruned data in the form of a multi-branch tree structure, wherein the first parameter indicates whether to prune the node based on the contribution of the node in the multi-branch tree structure, and the second parameter indicates whether to prune the node based on the importance of the node in the multi-branch tree structure. The method also includes normalizing and serializing the pruned data in the form of a multi-branch tree structure to generate normalized serialized data, wherein the normalized serialized data is presented in the form of a text description. In addition, the method also includes generating an attribution summary for the data based on the normalized serialized data by a language model.

[0006] According to a second aspect of the present disclosure, a device for data attribution is provided. The device includes a data acquisition module configured to acquire data in the form of text description, wherein the data includes changes in the data within a date range. The device includes an analysis tree generation module configured to generate an analysis tree based on the data in the form of text description and preset dimensions, wherein the analysis tree includes at least one of node attribution data and whole tree attribution data, the node attribution data indicating that the data is presented in the form of node hierarchical nesting, each node in the node attribution data includes a key numerical description of the node, and the whole tree attribution data is determined based on the node's disassembly path. The device also includes an analysis tree pruning module configured to prune the analysis tree based on a first parameter and a second parameter to obtain data in the form of a pruned multi-branch tree structure, wherein the first parameter indicates whether to prune the node based on the contribution of the node in the multi-branch tree structure, and the second parameter indicates whether to prune the node based on the importance of the node in the multi-branch tree structure. The device also includes a data generation module configured to normalize and serialize the pruned data in the form of a multi-branch tree structure to generate normalized serialized data, wherein the normalized serialized data is presented in the form of a text description. In addition, the device also includes an attribution summary generation module configured to generate an attribution summary for the data based on the normalized serialized data by a language model.

[0007] According to a third aspect of the present disclosure, an electronic device is provided, comprising a processor and a memory coupled to the processor, wherein the memory has instructions stored therein, and when the instructions are executed by the processor, the electronic device executes the method according to the first aspect.

[0008] In a fourth aspect of the present disclosure, a computer program product is provided. The computer program product is tangibly stored on a non-transitory computer readable medium and comprises computer executable instructions, which when executed cause a computer to perform the steps of the method of the first aspect of the present disclosure.

[0009] The purpose of this Summary is to introduce a selection of concepts in a simplified form that are further described in the Detailed Description below. This Summary is not intended to identify key features or essential features of the claimed subject matter, nor is it intended to limit the scope of the claimed subject matter. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] The above and other features, advantages and aspects of the embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. In the accompanying drawings, the same or similar reference numerals represent the same or similar elements, wherein:

[0011] Figures 1A-1EA schematic diagram showing an example environment in which devices and / or methods according to embodiments of the present disclosure may be implemented;

[0012] Figure 2 A flow chart of a method for data attribution according to an embodiment of the present disclosure is shown;

[0013] Figure 3 A schematic diagram showing an example framework for data attribution according to an embodiment of the present disclosure is shown;

[0014] Figure 4 A schematic diagram showing another example system framework for data attribution according to an embodiment of the present disclosure is shown;

[0015] Figure 5A A schematic diagram showing an example structure of rule-based data attribution according to an embodiment of the present disclosure;

[0016] Figure 5B A schematic diagram showing how to convert a text description into a multi-branch tree structure according to an embodiment of the present disclosure

[0017] Figure 5C An example schematic diagram of determining the balance of direct contributions of nodes in a candidate path according to an embodiment of the present disclosure is shown;

[0018] Figure 5D A schematic diagram showing an example of performing importance clipping according to an embodiment of the present disclosure;

[0019] Fig. 6A A schematic diagram showing an example structure of model-based data attribution according to an embodiment of the present disclosure;

[0020] Figure 6B-6D A schematic diagram showing a form of training data generated for training according to an embodiment of the present disclosure;

[0021] Fig. 6E A schematic diagram showing example prompts for training generated according to an embodiment of the present disclosure;

[0022] Figure 7 A block diagram showing an apparatus for data attribution according to some embodiments of the present disclosure; and

[0023] Figure 8 A block diagram of an electronic device according to an embodiment of the present disclosure is shown.

[0024] Throughout the drawings, the same or similar reference numbers denote the same or similar elements. DETAILED DESCRIPTION

[0025] It is understandable that before using the technical solutions disclosed in the embodiments of the present disclosure, the type, scope of use, usage scenarios, etc. of the personal information (such as voice) involved in the present disclosure should be informed to the user and the user's authorization should be obtained in an appropriate manner in accordance with relevant laws and regulations.

[0026] Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as being limited to the embodiments described herein, which are instead provided for a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not intended to limit the scope of protection of the present disclosure.

[0027] In the description of the embodiments of the present disclosure, the term "including" and similar terms should be understood as open inclusion, that is, "including but not limited to". The term "based on" should be understood as "based at least in part on". The term "one embodiment" or "the embodiment" should be understood as "at least one embodiment". The terms "first", "second", etc. may refer to different or the same objects, unless explicitly stated. Other explicit and implicit definitions may also be included below.

[0028] As mentioned above, the importance of data analysis is constantly highlighted, and the role of data attribution, as an important part of data analysis, is also becoming increasingly significant. Taking a key indicator in the data as an example, when the key indicator changes, the user needs to analyze the reason for the change of the key indicator, that is, the user expects to find out the main factors that cause the change of the indicator. For example, for key indicator A, sub-indicators X and Y can be decomposed according to dimension B, and sub-indicators P and Q can be decomposed according to dimension C, and then decomposed downward in order to find the main factors affecting the change of key indicator A. This decomposition path can be called a decomposition dimension. The depth and dimension of the decomposition can be determined by the user himself according to the user's demand for data attribution.

[0029] At present, one solution to data attribution is that users enumerate the factors that cause the change of the key indicator according to the decomposition dimension, and analyze them one by one in order to find the main factors. However, this manual analysis method is not only time-consuming and labor-intensive, but also has a high error rate. In order to avoid manual errors, some related technologies use rule-based methods for automatic attribution, which can achieve accurate attribution. However, this method is not suitable for the attribution of data for businesses that are frequently iteratively developed because it lacks flexibility and cannot adapt to business rules and changing structures. Some other related technologies use deep learning-based models to achieve automatic data attribution, but this model-based method is not suitable for situations where the amount of data is large and continues to grow. Because when the amount of data is extremely large, the model often makes mistakes when attributing data due to the current level of technology.

[0030] To this end, at least in order to solve the above problems and potential problems, an embodiment of the present disclosure proposes a solution for automatically attributing data to users. The solution first obtains data in the form of text description, wherein the data includes changes in the data within a date range, and based on the data in the form of text description and preset dimensions, generates an analysis tree including at least one of node attribution data and whole tree attribution data. Node attribution data is a data form presented in the form of node nesting, and each node in the node attribution data includes a description of the key value of the node, while the whole tree attribution data can be determined according to the node disassembly path. With the help of data in the analysis tree structure, the parent-child relationship and brother relationship between the data can be clearly expressed, which enhances the readability of data processing. Then, the analysis tree can be pruned according to the first parameter indicating the contribution of the node and the second parameter indicating the importance of the node to obtain the pruned data in the form of a multi-branch tree structure, so that the subsequent steps can focus on the nodes that have a greater influence on the data, while also avoiding the situation where the subsequent model is not general enough for data processing. Then, the pruned data in the multi-tree structure is normalized and serialized, so that the data is described in the form of text, which can improve the efficiency of subsequent data processing by the language model. Finally, the language model generates an attribution summary based on the normalized and serialized data in the text form.

[0031] Therefore, according to the solution of the embodiment of the present disclosure, by generating analysis tree structured data and combining multiple processing steps (including clipping, normalization and serialization, etc.), the accuracy and pertinence of data attribution are improved. At the same time, the attribution summary is generated using the language model to effectively extract key information. As a result, users can efficiently and accurately obtain data attribution results when facing complex data, greatly improving the user's operational experience in data attribution.

[0032] The embodiments of the present disclosure will be described in detail below in conjunction with the accompanying drawings. Figure 1A 1 is a schematic diagram of an example environment 100 in which devices and / or methods according to embodiments of the present disclosure may be implemented. Figure 1A As shown, the text description 110 shows the text description of the change details, change value and change rate of the key indicator A in the time range. When the key indicator A changes, the user needs to summarize the possible reasons affecting the change of the key indicator A, so as to make decisions on the business related to the key indicator A. In other words, attribution requires the output of the interpretation of the main factors causing the change of the key indicator A.

[0033] However, due to the different division of labor among users, some users lack a deep understanding of the relevant business and are unable to accurately analyze the reasons for the changes in key indicator A. Therefore, in order to help some users understand the reasons for changes in business indicators and make this attribution work a common task, data attribution can be automatically achieved by combining rule-based processing methods with model-based processing methods. This method that combines rule and model processing can ensure the stability of data attribution results.

[0034] refer to Figure 1A In order to dig deeper into the reasons that affect the changes in key indicator A, a tree-structured data representation method can be used to deconstruct key indicator A. The reason is that the tree structure can display data in a hierarchical manner. This hierarchical structure can clearly show the parent-child relationship, brother relationship, etc. between the data, which is convenient for the subsequent processing steps to quickly locate the nodes with great influence and their related paths. For example, the text description 110 about key indicator A (data 120 in the form of a text description of the changes in the key indicator) can be converted into a multi-branch tree structure 130, and then the next step of processing is performed on the data of the tree structure. It can be understood that key indicator A can be a root indicator or one of the sub-indicators under the root indicator.

[0035] In some embodiments, data in the form of text description may be converted into an analysis tree according to preset dimensions. Figure 1B FIG. 1 is a schematic diagram showing an example disassembly path 100B according to an embodiment of the present disclosure. Figure 1B, the key indicator A can be decomposed into sub-indicators X and Y according to the A dimension, and can also be decomposed into sub-indicators M and N according to the B dimension. For sub-indicators M and N, they can be decomposed into sub-indicators P and Q according to the C dimension. Similarly, sub-indicators can be further decomposed layer by layer according to other dimensions. The preset dimension here can be a decomposition dimension or decomposition path pre-defined by the user. Taking the key indicator A as the order rate as an example, there are multiple decomposition dimensions for the order rate. For example, according to the industry category dimension, the order rate can be divided into multiple sub-indicators such as the beauty order rate, the watch accessories order rate, and the 3C digital home appliance order rate. In addition, it can also be decomposed according to the business identity dimension. The order rate can be decomposed into multiple sub-indicators such as the cross-border order rate, the next-day order rate, and the instant retail order rate. The decomposition dimension (path) can be freely selected by the user according to their own needs. For the convenience of description, the (root) node and the (key) indicator can refer to each other in this article.

[0036] In some embodiments, the generated analysis tree may be in at least one of a node attribution data form or a whole tree attribution data form. Figure 1C FIG. 1 is a schematic diagram showing an example node attribution data form 100C according to an embodiment of the present disclosure. Figure 1C , the node attribution data is presented in the form of node hierarchical nesting, and each node in the node attribution data includes a key numerical description of the node. Node attribution needs to find the node (sub-indicator) with the largest contribution rate to the root indicator under the current decomposition dimension. It can be understood that the size of the contribution can be measured by the contribution rate data.

[0037] Figure 1D FIG. 1 shows a schematic diagram of an example root node 100D according to an embodiment of the present disclosure. Figure 1D As shown in the figure, the root node includes the key numerical description of the sub-indicator, possible trends, possible abnormal points and other text descriptions. Figure 1A The text description 110 shown is identical.

[0038] Figure 1E FIG. 1 is a schematic diagram of an example whole tree attribution data form 100E according to an embodiment of the present disclosure. Figure 1E The whole tree attribution data includes a disassembly path consisting of a root node and multiple disassembly nodes, and may also include the disassembly of possible multiplication factors. Usually, there is only one level of disassembly of possible multiplication formulas. Figure 1BTo describe, for example, for the key indicator Q, its possible multiplication factor decomposition can be R*S*T, because Q=R*S*T. The whole tree attribution data is determined based on the node decomposition path. Whole tree attribution needs to first determine which dimension decomposition method is the most appropriate, and then determine the node with the largest contribution to the root indicator from the decomposition method according to the node attribution method. It can be understood that the contribution rate value is the proportion of the contribution of a certain part to the overall change in the whole, that is, the proportion of the contribution of the sub-indicator to the change of the root indicator in the whole.

[0039] In some embodiments, in order to ensure the accuracy of subsequent language model processing of data, the analysis tree can also be adaptively pruned. That is, the text description of the analysis tree is converted into a standard multi-branch tree structure that can distinguish the importance of tree nodes at the same level to achieve node pruning. This can not only remove the impact of the text format of the data on the model calculation, but also screen out the most important sub-indicators to improve the efficiency of model calculation.

[0040] In some embodiments, for the sake of data processing efficiency, it is possible to determine whether to prune a node based on a first parameter indicating the contribution of a node in a multi-branch tree structure, because a node or sub-indicator with a small contribution cannot be the main cause of the change in the root indicator, and therefore can be pruned. In some embodiments, it is also possible to determine whether to prune a node based on a second parameter indicating the importance of a node in a multi-branch tree structure, because when the depth of the multi-branch tree structure is deep, the model may make mistakes when processing related data, so in order to avoid unnecessary mistakes, some nodes can be pruned based on the importance of the nodes.

[0041] Back to Figure 1A In some embodiments, in order to further avoid the influence of data format on model calculation, the data of the multi-tree structure 130 can be normalized and serialized so that it is presented as normalized and serialized text data 140. For example, the description with format in the data can be replaced with a universal descriptor so that the subsequent model can focus on the specific logic rather than the format in the process of attributing the data, which can improve the efficiency of model output. In some embodiments, before obtaining the normalized and serialized text data 140, the data of the multi-tree structure 130 needs to be output as a text description.

[0042] In some embodiments, after the relevant data of the key indicator A is processed according to the processing rules, the attribution summary 160 of the key indicator A can be obtained with the help of the language model 150. In some embodiments, the language model is a language model built based on deep learning technology, which has a large number of parameters and powerful language processing capabilities.

[0043] Combination Figure 1B In some embodiments, when the user wants to view a certain decomposition dimension for key indicator A, the user can click on the corresponding node, for example, click on the "decomposition by dimension B" control 131, and the main factors causing the change of key indicator A can be obtained by decomposing according to the dimension, and a series of detailed statistical indicator information 132 can also be attached. In some embodiments, when the user wants to pay attention to the reasons for the overall change of indicator A, the user can click on a button, so that the overall change of key indicator A and the most critical change factors from a global perspective can be displayed to the user.

[0044] Therefore, combined with Figures 1A-1E According to the solution provided by the embodiment of the present disclosure, the data is first structured by generating an analysis tree, and combined with processing steps such as clipping, normalization and serialization, the accuracy and pertinence of data attribution are significantly enhanced. In this process, the key information is efficiently refined by generating an attribution summary with the help of a language model. In this way, through the solution of the embodiment of the present disclosure, the results of data attribution can be obtained quickly and accurately, and the user experience in the data attribution operation process is comprehensively improved, allowing users to complete tasks more conveniently and efficiently in data attribution work, freeing themselves from complex data processing and focusing on making decisions based on attribution results.

[0045] It should be understood that the architecture and functions in the example environments 100-100E are described for exemplary purposes only, and do not imply any limitation on the scope of the present disclosure. Embodiments of the present disclosure may also be applied to other environments with different structures and / or functions.

[0046] The following will combine Figures 2 to 8 The process according to the embodiment of the present disclosure is described in detail. For ease of understanding, the specific data mentioned in the following description are exemplary and are not intended to limit the scope of protection of the present disclosure. It is understood that the embodiments described below may also include additional actions not shown and / or the actions shown may be omitted, and the scope of the present disclosure is not limited in this respect.

[0047] Figure 2 A flow chart of a method 200 for data attribution according to an embodiment of the present disclosure is shown. Next, the method 200 is schematically described by taking a device for data attribution as an execution subject as an example. The method 200 includes a block 202, a block 204, a block 206, a block 208, and a block 210.

[0048] At block 202, data in the form of a text description may be obtained, wherein the data includes changes in the data within a date range. Figure 1AAs shown, the text description 110 shows the text description of the change details, change value and change rate of the key indicator A in the time range. When the key indicator A changes, the user needs to summarize the possible reasons affecting the change of the key indicator A for decision-making on the business related to the key indicator A.

[0049] At block 204, an analysis tree may be generated based on the data in the form of text description and the preset dimensions, wherein the analysis tree includes at least one of node attribution data and whole tree attribution data, the node attribution data indicating that the data is presented in the form of node hierarchical nesting, each node in the node attribution data includes a key numerical description of the node, and the whole tree attribution data is determined based on the node disassembly path. Figure 1A In order to deeply explore the reasons that affect the changes in key indicator A, a tree-structured data representation method can be used to deconstruct key indicator A. In some embodiments, the data in the form of text description can be converted into an analysis tree according to preset dimensions. Figure 1B , the key indicator A can be decomposed into sub-indicators X and Y according to the A dimension, and can also be decomposed into sub-indicators M and N according to the B dimension. For sub-indicators M and N, they can be decomposed downward into sub-indicators P and Q according to the C dimension. Similarly, sub-indicators can continue to be decomposed layer by layer according to other dimensions. The preset dimension here can be a decomposition dimension or decomposition path predefined by the user. In some embodiments, the generated analysis tree can be in the form of node attribution data or at least one of the forms of whole tree attribution data. Combined with Figure 1C , the node attribution data is presented in the form of node hierarchical nesting, and each node in the node attribution data includes a key numerical description of the node. Node attribution needs to find the node (sub-indicator) with the largest contribution rate to the root indicator under the current decomposition dimension. Figure 1E The whole tree attribution data includes a disassembly path consisting of a root node and multiple disassembly nodes. The whole tree attribution data is determined based on the disassembly path of the node.

[0050] At box 206, the analysis tree can be pruned based on the first parameter and the second parameter to obtain data in the form of a pruned multi-branch tree structure, wherein the first parameter indicates whether to pruned the node based on the contribution of the node in the multi-branch tree structure, and the second parameter indicates whether to pruned the node based on the importance of the node in the multi-branch tree structure. In some embodiments, in order to ensure the accuracy of the subsequent language model processing of the data, the analysis tree can also be adaptively pruned. That is, the text description of the analysis tree is converted into a standard multi-branch tree structure that can distinguish the importance of tree nodes between the same levels to achieve node pruning. This can remove the influence of the text format of the data on the model calculation on the one hand, and can also filter out the most important sub-indicators to improve the efficiency of the model calculation. In some embodiments, for the efficiency of data processing, it can be determined whether to pruned a node based on the first parameter indicating the contribution of the node in the multi-branch tree structure, because the node or sub-indicator with a small contribution cannot be the main reason for the change of the root indicator, so the node can be pruned. In some embodiments, whether to prune a node can also be determined based on a second parameter indicating the importance of the node in the multi-tree structure. This is because when the depth of the multi-tree structure is deep, the model may make mistakes when processing related data. Therefore, in order to avoid unnecessary errors, some nodes can be pruned based on the importance of the nodes.

[0051] At block 208, the pruned data in the multitree structure form may be normalized and serialized to generate normalized serialized data, wherein the normalized serialized data is presented in a text description form. Figure 1A In some embodiments, in order to further avoid the influence of data format on model calculation, the data of the multi-branch tree structure 130 can be normalized and serialized so that it is presented as normalized and serialized text data 140. For example, the description with format in the data can be replaced with a universal descriptor so that the subsequent model can focus on the specific logic rather than the format in the process of attributing the data. This normalized and serialized method can improve the efficiency of model output. In some embodiments, before obtaining the normalized and serialized text data 140, the data of the multi-branch tree structure 130 needs to be output as a text description.

[0052] At block 210, a language model may generate an attribution summary for the data based on the normalized serialized data. Figure 1A In some embodiments, after the relevant data of the key indicator A is processed according to the processing rules, the attribution summary 160 of the key indicator A can be obtained with the help of the language model 150. In some embodiments, the language model is a language model built based on deep learning technology, which has a large number of parameters and powerful language processing capabilities.

[0053] Therefore, according to the solution of the embodiment of the present disclosure, by generating analysis tree structured data and combining multiple processing steps (including clipping, normalization and serialization, etc.), the accuracy and pertinence of data attribution are improved. At the same time, the attribution summary is generated using the language model to effectively extract key information. As a result, users can efficiently and accurately obtain data attribution results when facing complex data, greatly improving the user's operational experience in data attribution.

[0054] Figure 3 FIG. 3 is a schematic diagram of an example framework 300 for data attribution according to an embodiment of the present disclosure. Figure 3 As shown, the example framework 300 includes a data pre-processing step 310, a model prediction step 320, and a data post-processing step 330. Among them, the data pre-processing step 310 also includes an adaptive clipping step 312 and a normalized serialization step 313. The model prediction step 320 includes a model reasoning step 322. In addition, the data post-processing step 330 includes a normalized deserialization step 332. With the help of the data pre-processing step 310, a general data form suitable for model reasoning can be obtained, so that the model used for reasoning can focus on the logic of specific data attribution and improve the efficiency of model output. After obtaining the output result of the model reasoning, it is necessary to further perform a normalized deserialization step 332 on the result of the model output, so that the attribution summary of the model output can be presented in a user-readable form.

[0055] In some embodiments, after the original data passes through the adaptive clipping step 312 and the normalized serialization step 313, it can obtain a data form suitable for the model reasoning step 322, and then after the model reasoning step 322 and the normalized deserialization 332, it can obtain a readable and refined final data attribution result for the original data. Through this method of combining data processing rules and model reasoning, it can not only ensure the stability of data attribution for data, but also ensure that the model has good generalization output, thereby improving the user experience of data attribution.

[0056] Figure 4 A schematic diagram of another example system framework 400 for data attribution according to an embodiment of the present disclosure is shown. The example system framework 400 mainly includes two modules: data processing 410 and model building 420. The data processing 410 module is responsible for optimizing the input of the model, including adaptive clipping 402, normalized serialization 403, and normalized deserialization 405. It can be understood that there is a sequential dependency between adaptive clipping 402, normalized serialization 403, and normalized deserialization 405. The model building 420 is responsible for training the model for reasoning, including training data construction 422, model training 424, and model testing 426. Figure 3, Figure 3 The steps of data pre-processing step 310 and data post-processing step 330 can be Figure 4 The steps performed by the data processing 410 module are shown. Figure 5A as well as Fig. 6A The detailed steps that can be performed by the two modules of data processing 410 and model building 420 are further described. Figure 5A A schematic diagram of an example structure 500A for rule-based data attribution according to an embodiment of the present disclosure is shown. Fig. 6A A schematic diagram of an example structure 600A of model-based data attribution according to an embodiment of the present disclosure is shown. Figures 5A-6A To further describe the example system framework 400. For ease of description, the following model may refer to a large language model based on deep learning.

[0057] refer to Figure 4 In the data processing module 410, adaptive cropping 402 may be performed on the original data 401. In some embodiments, the original data 401 may be as follows: Figure 1B-1E The data in the form of node attribution shown or the data in the form of whole tree attribution. Figure 5A When performing adaptive clipping 402, data structuring 501 may be performed first, that is, the original data 401 may be converted from the text description into a standard multi-tree structure. Through the data structure 501, the input content for model reasoning can be refined, and the influence of the text format on the model reasoning can also be removed. In some embodiments, the levels of the multi-tree are distinguished by the indentation of the text description, and the sub-indicator with less indentation represents a higher level in the multi-tree structure. Figure 5B FIG. 500B is a schematic diagram showing an example of converting a text description into a multi-branch tree structure according to an embodiment of the present disclosure. Figure 5B , the multi-branch tree structure includes multiple nodes, each of which corresponds to a sub-index. Figure 5B The method of converting the text description into a standard multi-branch tree structure can clearly distinguish the importance of tree nodes at the same level.

[0058] Continue to refer Figure 4In order to improve the speed of model reasoning, the input required for model reasoning can be simplified by contribution clipping 502. Contribution refers to the proportion of the contribution of each child node in the multi-branch tree structure to the change of the root node in the whole (which can be measured by the contribution rate), so contribution clipping 502 is for the child nodes in the multi-branch tree structure. In view of the fact that the child nodes with small contribution are not the main influencing factors of the root node change, the input of model reasoning can be simplified by clipping the child nodes with low contribution. In some embodiments, it can be determined whether to clip a node by comparing the relationship between the contribution of each node and the first parameter used to indicate the contribution. Furthermore, in view of the fact that the multi-branch tree structure can be divided into node attribution data form and whole tree attribution data form, the execution of contribution clipping 502 can also be divided into two sets of logic.

[0059] In some embodiments, when the multi-branch tree structure is data in the form of node attribution, given that the nodes in the multi-branch tree structure are disassembled according to a certain dimension to the parent indicator, the logic of cutting the contribution of node attribution is that in the same level, it is necessary to find the node with the maximum contribution rate, and determine its contribution as the maximum contribution rate, and compare the contribution rate of all other nodes in the level with the maximum contribution rate. If the ratio of the current node's contribution rate to the maximum contribution rate is less than the first parameter, then the node will be cut; if the ratio of the current node's contribution rate to the maximum contribution rate is greater than or equal to the first parameter, then the node can be retained. In some embodiments, the first parameter for indicating the contribution of the node can be used as a hyperparameter and configured in the process of model training and reasoning. In some embodiments, for all nodes in the multi-branch tree, it can be determined whether it is necessary to cut certain nodes in each level through traversal by standardized algorithms such as top-k and same-layer information.

[0060] In some embodiments, when the multi-branch tree structure is data in the form of whole tree attribution, given that the focus of the whole tree attribution form is to screen out the most important disassembly direction and the node with the largest contribution in the whole tree, it is not limited to a specific disassembly dimension. Therefore, when performing contribution clipping 502, it is necessary to select the disassembly path. Taking key indicator A as an example, key indicator A can be disassembled according to the A dimension or the B dimension. When the contribution rates of the four sub-indicators disassembled according to the A dimension are all 25%, and one of the four sub-indicators disassembled according to the B dimension has a contribution rate of 70%, then the disassembly path disassembled according to the B dimension is a disassembly path that can highlight the key points. At this time, the disassembly according to the B dimension is a better disassembly path than the disassembly according to the A dimension, because the disassembly according to the B dimension can find a factor that has the greatest impact on key indicator A, so that when the user finds the main factors affecting key indicator A, it can realize the subsequent adjustment of the response strategy for business changes related to key indicator A.

[0061] Therefore, in some embodiments, when performing contribution pruning for the entire tree, it is necessary to traverse the contribution rates of all leaf nodes to the root node, and then select K (number of candidates) disassembly paths as candidate paths. It is understandable that the paths selected at this time are all paths that completely reach the leaf nodes. In the process of selecting candidate paths, whether the path is selectable can be determined based on the balance of the direct contribution rates of the child nodes under each candidate path. Combined with Figure 5C , Figure 5C A schematic diagram of an example 500C of determining a balance degree of direct contributions of nodes in a candidate path according to an embodiment of the present disclosure is shown. Figure 5C The Gini coefficient is used to characterize the degree of balance of the direct contribution rate of the sub-nodes under the disassembly path. If the Gini coefficient is closer to 0, it means that the degree of balance of the direct contribution rate of the sub-nodes under the disassembly path is more uneven; if the Gini coefficient is closer to 1, it means that the degree of balance of the direct contribution rate of the sub-nodes under the disassembly path is more even. The calculation formula of the Gini coefficient is as follows: Among them, Figure 5C As shown, A is the area of ​​the upper half of the curve, and B is the area of ​​the lower half of the curve.

[0062] To describe more clearly Figure 5CThe Gini coefficient is used to characterize the degree of balance of the direct contribution rate of the sub-nodes under the disassembly path. The following example is used to describe it: Assuming that under a certain disassembly path, the contribution rates of the sub-nodes X, Y, and Z are 10%, 20%, and 70%, respectively; the ratios of the change values ​​of the sub-nodes X, Y, and Z to the root node are 30%, 40%, and 30%, respectively. Then we can first draw a curve graph according to the contribution rate in ascending order. At this time, the area under the X curve is (10%+0) * 30% / 2 = 0.015. Similarly, the area under the Y curve is (10%+30%) * 40% / 2 = 0.08, and the area under the Z curve is (30%+100%) * 30% / 2 = 0.195. Therefore, the area of ​​B is the sum of the three parts, 0.29, and the area of ​​A is 0.21. At this time, according to formula (1), we can calculate G=0.21 / 0.5=0.42, which is close to 0, indicating that the contribution rates of the sub-nodes under this decomposition path are still different.

[0063] In some embodiments, after the K candidate paths for the whole tree attribution are determined, it is also necessary to determine whether the disassembly depth D of these paths is necessary. Figure 5B , the depth of the key indicator A of the root node is 0. The depth of node P is 2. If the disassembly depth D of the candidate path is greater than the preset disassembly depth, then all candidate paths with a depth greater than D can be discarded. In some embodiments, the disassembly depth D and the candidate number K of the candidate path can be used as hyperparameters, i.e., the first parameters, for pruning the contribution rate of the whole tree attribution. Users can choose different first parameters according to different types of businesses related to the root node. In summary, in the process of pruning the contribution of the whole tree attribution, it is necessary to consider the depth D of the disassembly path and the candidate number K of the candidate path.

[0064] refer to Figure 5A After executing contribution clipping 502, in order to further ensure that the logical ability of model processing is in a normal state, importance clipping 503 can be performed. The reason is that after contribution clipping 502, the structure of the multi-branch tree may not be within the scope of model processing. Therefore, importance clipping 503 can be used as a backup operation to ensure the availability of model input content.

[0065] In some embodiments, after performing contribution pruning 502 on the multi-tree, the contribution between the levels of the multi-tree can be reordered so that the nodes with large contribution in the same level are ordered at the top (left), and then the importance pruning 503 of the loop is performed. In some embodiments, pruning can be started from the subsequent traversal until the depth of the multi-tree meets the requirement (or until the number of lines of the text description meets the requirement of the second parameter for indicating the importance of the node).

[0066] In some embodiments, in view of the difference between the whole tree attribution and the node attribution, the strength of executing the importance clipping 503 is also different. In the node attribution, the minimum strength of the importance clipping of the node attribution is the non-leaf node. The reason is that when the user clicks on a certain node and wants to view the details of the contribution rate of each child node under the disassembly dimension, if the child node is clipped, then after the child node is clipped, only the contribution rate of the parent node of the child node will be left. At this time, the user will want to see the details of the contribution rate of the child node, but there will be no child nodes to see due to the clipping. Therefore, when executing the importance clipping of the node attribution, only the non-leaf nodes can be clipped, so that the parent node with a lower contribution rate in the parent node of the child node can be found, and the parent node can be deleted together, while the parent node with a higher contribution rate can retain the disassembly of its child node. In the whole tree attribution, the strength of executing the importance clipping 503 is not limited. Figure 5D FIG. 5 is a schematic diagram showing an example 500D of performing importance clipping according to an embodiment of the present disclosure. Figure 5D When performing importance pruning of node-attributed data, M, P, and Q in the solid-line frame can be deleted together; and when performing importance pruning of the entire tree-attributed data, P in the dashed frame can be deleted.

[0067] Continue to refer Figure 5A After the contribution clipping 502 and the importance clipping 503 are performed on the original data 501, the structured data textualization 504 can be performed. In some embodiments, the structured data textualization is to output the data in the form of a multi-branch tree structure as data in the form of a text description. In some embodiments, in order to ensure the integrity of the data, a depth-first traversal (DFS) algorithm can be used to output the data in the form of a text description.

[0068] Back to Figure 4 After the adaptive pruning 402 is performed, the normalization serialization 403 may be performed to enable the model to obtain input focused on logic. Figure 5A , the normalization serialization 403 can be divided into template matching 505 and key description replacement 506. In some embodiments, the template matching 505 is performed to extract key information of a specific style in the text description. For example, for the following description, the information of multiple fields such as the disassembly dimension "[disassembly dimension_1]"; the indicators are [sub-indicator_2] and [sub-indicator_3]; and the key value description can be obtained.

[0069]

Disassembly dimension_1

Sub-indicator_2

Sub-indicator_3

[0070] In some embodiments, regular expressions may be used to implement template matching 505 of key information. In some embodiments, after the template matching 505 operation of key information is completed, key description replacement 506 may be performed, that is, the disassembled dimensions, indicators, and key values ​​obtained through template matching 505 are replaced. In some embodiments, preset fixed symbols such as token_1, token_2, etc. may be used to replace key fields, which can reduce the number of format characters, and can also accelerate the model's understanding of the input fields and improve the efficiency of the model output. At the same time, the method of replacing key fields with fixed symbols can also prevent data leakage.

[0071] Back to Figure 4 After the normalization and serialization 403 is completed, the model (inference) 404 can be executed, as shown in Figure 5A Model reasoning 507 is shown. In some embodiments, when the data-processed text is input into the language model for reasoning, an output with fixed symbols can be obtained. At this time, it is also necessary to perform Figure 4 The normalized deserialization is shown as 405. Figure 5A , that is, key description restoration 508 can be performed. In some embodiments, key description restoration 508 refers to restoring the key description replaced by fixed symbols in the normalization serialization 403 operation (including template matching 505 and key description replacement 506) to the original description, so that the user can get a readable attribution summary of the original data 401. This method uses adaptive clipping in the rule part to streamline the information input to the model, and supplements the logical capabilities of the large model through partial rule judgment. At the same time, the process of normalization serialization, model reasoning and normalization deserialization is adopted to reduce the interference of data format on logic and reduce the time required for large model answers.

[0072] Therefore, by generating analysis tree structured data and combining multiple data processing steps (including pruning, normalization and serialization, etc.), the accuracy and pertinence of data attribution are improved. At the same time, the attribution summary is generated using the language model to effectively extract key information. As a result, users can efficiently and accurately obtain data attribution results when faced with complex data, greatly improving the user's operational experience in data attribution.

[0073] Continue to refer Figure 4 As described above, data processing 410 is responsible for optimizing the input of the model, while model building 420 is responsible for training the model for reasoning, which includes training data construction 422, model training 424, and model testing 426. It can be understood that there is a sequential dependency between training data construction 422, model training 424, and model testing 426.

[0074] The following will be combined Fig. 6A Model building 420 is further described. Fig. 6A FIG. 6 is a schematic diagram showing an example structure 600A of model-based data attribution according to an embodiment of the present disclosure. Fig. 6A As shown, before executing model training 424, training data construction 422 needs to be executed first, because the model needs to adjust the model parameters that determine the model behavior and performance according to the training data. In some embodiments, training data construction 422 can be further subdivided into forged data generation 601, data processing 602 and legitimacy verification 603.

[0075] In some embodiments, in view of the powerful text generation capability of the language model, training data may be generated based on a predetermined format using the language model. Figure 6B-6D That is, the predetermined format is described. In some embodiments, the training data includes training node attribution data and training whole tree attribution data. As mentioned above, the node attribution data and the whole tree attribution data are multiple types of multi-branch trees, so the purpose of generating training data is to generate multiple types of multi-branch tree structures.

[0076] In order to enhance the generalization of the model, in the process of generating training data, a fixed symbol similar to token_1 can be used to represent the key indicators described by each multi-branch tree node, and the key values ​​in the key value description can be randomly generated. In some embodiments, for the effectiveness of model training, the depth of the generated training multi-branch tree can be set to a maximum of 7, and the number of child nodes per layer can be 1-5. Figure 6B , Figure 6B A schematic diagram showing an example node attribution 600B generated for training according to an embodiment of the present disclosure is shown.

[0077] In some embodiments, the constructed training root node may include a time range (baseline period and observation period), a key data description of a key node (key indicator), anomalies detected during the observation period, a trend detected during the observation period, etc. In some embodiments, the date of the constructed training root node may randomly select any X days from January 1, 1900 to December 31, 2030 as the base period and the observation period, and it is also necessary to ensure that there is no date overlap between the base period and the observation period, and the observation period is after the base period. In some embodiments, the key data description of the key node (key indicator) in the constructed training root node may be generated in a random manner. In some embodiments, the type of anomalies in the constructed training root node may be set according to "the fluctuation exceeds 2 standard deviations of the average fluctuation of 30 days", "compared to the previous day, the month-on-month fluctuation exceeds 10.0%", and "the fluctuation exceeds 2 standard deviations of the average fluctuation of 7 days", and the training data of the root node may be constructed according to the permutation and combination of these three types of anomalies, and the probability of occurrence of the anomaly may also be set to 50% to generate the training data of the root node. In some embodiments, the probability of a trend occurring during the observation period in the constructed training root node may also be set to 50%, where the trend is an increase or decrease. Figure 6C , Figure 6C A schematic diagram showing an example root node 600C generated for training according to an embodiment of the present disclosure is shown.

[0078] In some embodiments, since the multi-branch tree structure in the form of a whole tree includes a root node, a disassembly node, and a formula node, when generating training whole tree data, the above process of generating training node attribution data and training root node data can be used as a reference. Formula nodes (such as sub-indicators A=B*C*D*E) are special disassembly nodes in the multi-branch tree structure in the form of a whole tree, but its hierarchical depth is 1 layer. Therefore, when generating training whole tree data, the process of generating disassembly nodes can also be used as a reference, but the depth of its multi-branch tree structure needs to be set to a maximum of 1. Reference Fig.6D , Fig.6D A schematic diagram showing an example whole tree 600D generated for training according to an embodiment of the present disclosure is shown.

[0079] Back to Fig. 6A In some embodiments, after the forged data generation 601 is performed with the help of the language model, data processing 602 needs to be performed, that is, obtaining the labels in the training data. In some embodiments, the labels in the training data can be generated with the help of data processing (model) such as a language model. Fig. 6E , Fig. 6E A schematic diagram showing an example prompt 600E generated for training according to an embodiment of the present disclosure is shown.

[0080] refer to Fig. 6AIn some embodiments, after the steps of generating and processing the training data are completed, in order to ensure that the generated training data does not affect the training of the model, it is also necessary to perform a legality verification 603. For example, during the training process, if the input description is "token_1 contribution rate<r 50.70%> **", and the model answer may be "token_1 contributes the most, with a contribution rate of<r 50.70> ", this incomplete reference is not allowed because<r 50.70%> The corresponding one is ** <font color="red">50.70%< / font> **, r means the font format is red. That is, if there is a missing symbol in the answer (such as a percent sign), the normalized deserialization will fail. Therefore, the legality check is mainly to verify whether the key values ​​in the model answer appear in the vocabulary. If the key values ​​that are not in the vocabulary appear in the answer, it means that the model does not fully reference the data format, and such data will be removed.

[0081] Back to Figure 4 After the training data construction 422 is completed, the model training 422 can be performed. Fig. 6A , model training 424 may include supervised fine-tuning 604, that is, the model may be trained using the SFT method. In some embodiments, specific parameters of the training process may be as shown in Table 1 below:

[0082] refer to Figure 4 After completing model training 424, the performance of the model needs to be tested, so model testing 426 needs to be performed. Fig. 6A In some embodiments, the model test 426 also needs to perform an evaluation standard test 605 and a secondary evaluation test 606. In some embodiments, the evaluation standard test 605 can be performed according to the rules of Table 2 below.

[0083]

[0084] In some embodiments, in order to ensure the full release of model performance, a secondary evaluation test 606 can also be performed on the model to achieve a comprehensive evaluation of the model. In some embodiments, real business data can be used to test the model, and the above 6 indicators can be used to comprehensively obtain the model score from 5 dimensions. In some embodiments, the model results can also be scored manually. Through this unsupervised method of automatically constructing training data, various types of training data for improving model performance can be generated. At the same time, a variety of evaluation and inspection methods are combined to ensure the effectiveness of model performance verification.

[0085] Therefore, according to the solution of the embodiment of the present disclosure, by generating analysis tree structured data and combining multiple processing steps (including clipping, normalization and serialization, etc.), the accuracy and pertinence of data attribution are improved. At the same time, the attribution summary is generated using the language model to effectively extract key information. As a result, users can efficiently and accurately obtain data attribution results when facing complex data, greatly improving the user's operational experience in data attribution.

[0086] Figure 7 1 shows a block diagram of an apparatus 700 for data attribution according to some embodiments of the present disclosure. Figure 7 As shown, the device 700 includes a data acquisition module 702, which is configured to acquire data in the form of text description, wherein the data includes changes in the data within a date range. The device 700 also includes a parsing tree generation module 704, which is configured to generate a parsing tree based on the data in the form of text description and a preset dimension, wherein the parsing tree includes at least one of node attribution data and whole tree attribution data, the node attribution data indicates that the data is presented in the form of node hierarchical nesting, each node in the node attribution data includes a key numerical description of the node, and the whole tree attribution data is determined based on the node's disassembly path. The device 700 also includes a parsing tree pruning module 706, which is configured to prune the parsing tree based on a first parameter and a second parameter to obtain data in the form of a pruned multi-branch tree structure, wherein the first parameter indicates whether to prune the node based on the contribution of the node in the multi-branch tree structure, and the second parameter indicates whether to prune the node based on the importance of the node in the multi-branch tree structure. The device 700 further includes a data generation module 708, which is configured to normalize and serialize the pruned data in the form of a multi-branch tree structure to generate normalized serialized data, wherein the normalized serialized data is presented in the form of a text description. In addition, the device 700 further includes an attribution summary generation module 710, which is configured to generate an attribution summary for the data based on the normalized serialized data by a language model.

[0087] Figure 8 A block diagram of an electronic device 800 is shown in accordance with certain embodiments of the present disclosure. Figure 8 8 is a block diagram of an electronic device 800 according to some embodiments of the present disclosure. The device 800 may be a device or apparatus described in an embodiment of the present disclosure. Figure 8As shown, the device 800 includes a central processing unit (CPU) and / or a graphics processing unit (GPU) 801, which can perform various appropriate actions and processes according to computer program instructions stored in a read-only memory (ROM) 802 or loaded from a storage unit 808 to a random access memory (RAM) 803. In the RAM 803, various programs and data required for the operation of the device 800 can also be stored. The CPU / GPU 801, the ROM 802, and the RAM 803 are connected to each other via a bus 804. An input / output (I / O) interface 805 is also connected to the bus 804. Although not shown in FIG. Figure 8 As shown in FIG. 8 , device 800 may also include a co-processor.

[0088] A number of components in the device 800 are connected to the I / O interface 805, including: an input unit 806, such as a keyboard, a mouse, etc.; an output unit 807, such as various types of displays, speakers, etc.; a storage unit 808, such as a disk, an optical disk, etc.; and a communication unit 809, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 809 allows the device 800 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0089] The various methods or processes described above may be performed by the CPU / GPU 801. For example, in some embodiments, the methods may be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage unit 808. In some embodiments, part or all of the computer program may be loaded and / or installed on the device 800 via the ROM 802 and / or the communication unit 809. When the computer program is loaded into the RAM 803 and executed by the CPU / GPU 801, one or more steps or actions in the methods or processes described above may be performed.

[0090] In some embodiments, the methods and processes described above may be implemented as a computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for executing various aspects of the present disclosure.

[0091] Computer readable storage medium can be a tangible device that can hold and store instructions used by an instruction execution device. Computer readable storage medium can be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination thereof. More specific examples (non-exhaustive list) of computer readable storage medium include: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a static random access memory (SRAM), a portable compact disk read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanical encoding device, for example, a punch card or a convex structure in a groove on which instructions are stored, and any suitable combination thereof. The computer readable storage medium used here is not interpreted as a transient signal itself, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagated by a waveguide or other transmission medium (for example, a light pulse by an optical fiber cable), or an electrical signal transmitted by a wire.

[0092] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to each computing / processing device, or downloaded to an external computer or external storage device via a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network can include copper transmission cables, optical fiber transmissions, wireless transmissions, routers, firewalls, switches, gateway computers, and / or edge servers. The network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards the computer-readable program instructions for storage in the computer-readable storage medium in each computing / processing device.

[0093] The computer program instructions for performing the operations of the present disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages, and conventional procedural programming languages. Computer-readable program instructions may be executed entirely on a user's computer, partially on a user's computer, as a separate software package, partially on a user's computer, partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., using an Internet service provider to connect via the Internet). In some embodiments, an electronic circuit, such as a programmable logic circuit, a field programmable gate array (FPGA), or a programmable logic array (PLA), may be customized by utilizing the state information of the computer-readable program instructions, and the electronic circuit may execute the computer-readable program instructions, thereby implementing various aspects of the present disclosure.

[0094] These computer-readable program instructions can be provided to a processing unit of a general-purpose computer, a special-purpose computer, or other programmable data processing device, thereby producing a machine, so that when these instructions are executed by the processing unit of the computer or other programmable data processing device, a device that implements the functions / actions specified in one or more boxes in the flowchart and / or block diagram is generated. These computer-readable program instructions can also be stored in a computer-readable storage medium, and these instructions cause the computer, programmable data processing device, and / or other equipment to work in a specific manner, so that the computer-readable medium storing the instructions includes a manufactured product, which includes instructions for implementing various aspects of the functions / actions specified in one or more boxes in the flowchart and / or block diagram.

[0095] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device so that a series of operating steps are performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to implement the functions / actions specified in one or more boxes in the flowchart and / or block diagram.

[0096] The flow chart and block diagram in the accompanying drawings show the possible architecture, function and operation of the equipment, method and computer program product according to multiple embodiments of the present disclosure. In this regard, each frame in the flow chart or block diagram can represent a part of a module, program segment or instruction, and a part of the module, program segment or instruction includes one or more executable instructions for realizing the specified logical function. In some alternative implementations, the function marked in the frame can also occur in a sequence different from that marked in the accompanying drawings. For example, two continuous frames can actually be executed substantially in parallel, and they can also be executed in the opposite order sometimes, depending on the functions involved. It should also be noted that each frame in the block diagram and / or flow chart, and the combination of frames in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.

[0097] The embodiments of the present disclosure have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and changes will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The choice of terms used herein is intended to best explain the principles of the embodiments, practical applications, or technical improvements to the technology in the market, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein.

[0098] Some example implementations of the present disclosure are listed below.

[0099] Example 1. A method for data attribution, comprising: Obtaining data in a text description format, the data including changes in the data within a date range; Based on the data in the text description form and the preset dimensions, an analysis tree is generated, the analysis tree including at least one of node attribution data and whole tree attribution data, the node attribution data indicating that the data is presented in a node hierarchical nested form, each node in the node attribution data including a key numerical description of the node, and the whole tree attribution data is determined based on a node disassembly path; Based on a first parameter and a second parameter, pruning the analysis tree to obtain pruned data in a multi-branch tree structure, wherein the first parameter indicates whether to prune the node based on the contribution of the node in the multi-branch tree structure, and the second parameter indicates whether to prune the node based on the importance of the node in the multi-branch tree structure; Normalizing and serializing the pruned data in the multi-tree structure form to generate normalized serialized data, wherein the normalized serialized data is presented in a text description form; and An attribution summary for the data is generated by a language model based on the normalized serialized data.

[0100] Example 2. The method according to Example 1, wherein based on the first parameter and the second parameter, pruning the analysis tree to obtain pruned data in the form of a multi-branch tree structure includes: Converting the analysis tree into data in a multi-branch tree structure, wherein the multi-branch tree structure includes a plurality of nodes; and Based on the first parameter and the second parameter, pruning nodes in the multi-tree structure to obtain pruned data in the multi-tree structure form; and The pruned data in the multi-tree structure form is converted into the pruned data in the multi-tree structure form presented in text form.

[0101] Example 3. According to the method of any one of Examples 1-2, the contribution is determined based on the contribution rate of the node, wherein based on the first parameter and the second parameter, pruning the nodes in the multi-tree structure to obtain the pruned data in the multi-tree structure form includes: In response to the analysis tree being node attribution data, traversing downward in sequence a plurality of nodes at each level from the level where the highest node of the data in the multi-branch tree structure corresponding to the node attribution data is located; Determine a maximum contribution rate, the maximum contribution rate indicating the contribution rate of a node having the largest contribution rate to the highest node among each node at the same level in the data in the multi-branch tree structure form corresponding to the node attribution data; Determine the ratio of the contribution rate of each node in each node of the same level in the data in the form of a multi-branch tree structure corresponding to the node attribution data to the maximum contribution rate; and In response to the ratio being less than the first parameter, in each node at the same level, each node whose ratio is less than the first parameter is pruned.

[0102] Example 4. The method according to any one of Examples 1-3, further comprising: In response to the analysis tree being adjusted based on the first parameter, sorting each node in the same level in the analysis tree; In response to the sorted analysis tree being node attribution data, determining the number of levels of each leaf node in the data in the multi-branch tree structure form corresponding to the node attribution data; and In response to the number of levels being greater than the second parameter, each node whose number of levels is greater than the second parameter and a previous non-leaf node of the node are pruned in a loop.

[0103] Example 5. According to any one of the methods of Examples 1-4, the first parameter includes a disassembly depth and a candidate number of a disassembly path, the disassembly depth and the candidate number are related to the disassembly path, wherein based on the first parameter and the second parameter, pruning the nodes in the multi-tree structure to obtain the pruned data in the form of the multi-tree structure includes: In response to the analysis tree being the whole tree attributed data, traversing each leaf node in the data in the multi-branch tree structure form corresponding to the whole tree attributed data; Determine the contribution rate of each leaf node to the root node in the data in the form of a multi-branch tree structure corresponding to the whole tree attribution data; and Based on the disassembly depth, the number of candidates, and the contribution rate of each leaf node to the root node in the multi-branch tree structure data corresponding to the whole tree attribution data, it is determined whether to prune each node.

[0104] Example 6. The method according to any one of Examples 1-5, further comprising: In response to the analysis tree being adjusted based on the first parameter, sorting each node in the same level in the analysis tree; Determine the number of levels of each leaf node in the sorted data in the multi-branch tree structure corresponding to the whole tree attribution data; and Based on the second parameter and the number of levels, it is determined in a loop whether to prune each leaf node in the data in the multi-branch tree structure form corresponding to the whole tree attribution data.

[0105] Example 7. The method according to any one of Examples 1-6, wherein converting the pruned data in the multi-tree structure form into the pruned data in the multi-tree structure form presented in text form comprises: Depth-first traversal of the pruned data in the multi-branch tree structure; and The pruned data in the multi-tree structure form is converted into the pruned data in the multi-tree structure form presented in text form.

[0106] Example 8. The method according to any one of Examples 1-7, wherein normalizing and serializing the pruned data in the multi-tree structure form to generate normalized serialized data comprises: Determine a plurality of fields in the data in the multi-tree structure form after the trimming presented in text form, wherein the plurality of fields include a key field and a data field; and The key field is replaced with a preset symbol to generate normalized serialized data.

[0107] Example 9. According to the method of any one of Examples 1-8, the language model is a first language model, wherein generating, by the language model based on the normalized serialized data, an attribution summary for the data comprises: The first language model generates an attribution summary with preset symbols based on the normalized serialized data; and The attribution summary with the preset symbol is normalized and deserialized to generate an attribution summary for the data.

[0108] Example 10. The method according to any one of Examples 1-9, further comprising: The second language model generates training data for the first language model based on a predetermined format, wherein the training data includes training node attribution data and training whole tree attribution data; and The first language model is trained based on the training data.

[0109] Example 11. The method according to any one of Examples 1-10, wherein generating, by the second language model based on a predetermined format, training data for the first language model comprises: The second language model generates a training label for the training data based on a preset prompt and the training data.

[0110] Example 12. The method according to any one of Examples 1-11, further comprising: Determining whether the generated training data satisfies a legality check; and In response to the generated training data not satisfying the legality check, the generated training data not satisfying the legality check is removed.

[0111] Example 13. The method according to any one of Examples 1-12, further comprising: In response to detecting a user interaction with the first control, displaying an attribution summary of node attribution data regarding the data; and In response to detecting a user interaction with the first control, an attribution summary of the entire tree of attribution data for the data is displayed.

[0112] Example 14. An apparatus for data attribution, comprising: A data acquisition module is configured to acquire data in a text description form, wherein the data includes changes in the data within a date range; A parse tree generation module is configured to generate a parse tree based on the data in the text description form and preset dimensions, the parse tree including at least one of node attribution data and whole tree attribution data, the node attribution data indicating that the data is presented in a node hierarchical nested form, each node in the node attribution data including a key numerical description of the node, and the whole tree attribution data being determined based on a node disassembly path; A parse tree pruning module is configured to prune the parse tree based on a first parameter and a second parameter to obtain pruned data in a multi-branch tree structure, wherein the first parameter indicates whether to prune the node based on the contribution of the node in the multi-branch tree structure, and the second parameter indicates whether to prune the node based on the importance of the node in the multi-branch tree structure; A data generation module is configured to normalize and serialize the pruned data in the multi-tree structure form to generate normalized serialized data, wherein the normalized serialized data is presented in a text description form; and The attribution summary generating module is configured to generate an attribution summary for the data based on the normalized serialized data by a language model.

[0113] Example 15. The apparatus of any one of Example 14, wherein the analysis tree pruning module comprises: A first conversion module is configured to convert the analysis tree into data in a multi-tree structure, wherein the multi-tree structure includes a plurality of nodes; and a first pruning module configured to prune nodes in the multi-tree structure based on the first parameter and the second parameter to obtain pruned data in the multi-tree structure; and The second conversion module is configured to convert the pruned data in the multi-tree structure form into the pruned data in the multi-tree structure form presented in text form.

[0114] Example 16. According to the apparatus of any one of Examples 14-15, the contribution is determined based on a contribution rate of a node, wherein the first trimming module comprises: A first traversal module is configured to, in response to the analysis tree being node attribution data, traverse downward in sequence a plurality of nodes at each level from the level where the highest node of the data in the multi-branch tree structure corresponding to the node attribution data is located; A first determination module is configured to determine a maximum contribution rate, wherein the maximum contribution rate indicates a contribution rate of a node having the largest contribution rate to the highest node among each node at the same level in the data in a multi-branch tree structure corresponding to the node attribution data; A second determination module is configured to determine a ratio of a contribution rate of each node in each node of the same level in the data in a multi-branch tree structure corresponding to the node attribution data to the maximum contribution rate; and The second pruning module is configured to, in response to the ratio being smaller than the first parameter, prune each node whose ratio is smaller than the first parameter in each node at the same level.

[0115] Example 17. The apparatus of any one of Examples 14-16 further includes: a first sorting module configured to sort each node in the same level of the analysis tree in response to the analysis tree being adjusted based on the first parameter; A third determination module is configured to determine the number of levels of each leaf node in the data in the multi-branch tree structure form corresponding to the node attribution data in response to the sorted analysis tree being node attribution data; and The third pruning module is configured to, in response to the number of levels being greater than the second parameter, cyclically prune each node whose number of levels is greater than the second parameter and the previous non-leaf node of the node.

[0116] Example 18. The apparatus according to any one of Examples 14-17, wherein the first parameter includes a disassembly depth and a candidate number of a disassembly path, the disassembly depth and the candidate number are related to the disassembly path, and wherein the first clipping module includes: A second traversal module is configured to traverse each leaf node in the data in a multi-branch tree structure corresponding to the whole tree attribution data in response to the analysis tree being the whole tree attribution data; a fourth determination module configured to determine a contribution rate of each leaf node to a root node in the data in a multi-branch tree structure corresponding to the whole tree attribution data; and The fifth determination module is configured to determine whether to prune each node based on the disassembly depth, the number of candidates, and the contribution rate of each leaf node to the root node in the multi-branch tree structure data corresponding to the whole tree attribution data.

[0117] Example 19. The apparatus of any one of Examples 14-18, further comprising: a second sorting module, configured to sort each node in the same level in the analysis tree in response to the analysis tree being adjusted based on the first parameter; a sixth determination module, configured to determine the number of levels of each leaf node in the sorted data in the multi-branch tree structure corresponding to the whole tree attribution data; and The seventh determination module is configured to cyclically determine whether to prune each leaf node in the data in the multi-branch tree structure form corresponding to the whole tree attribution data based on the second parameter and the number of levels.

[0118] Example 20. The apparatus of any one of Examples 14-19, wherein the second conversion module comprises: A third traversal module is configured to perform a depth-first traversal of the pruned data in the multi-branch tree structure; and The third conversion module is configured to convert the pruned data in the multi-tree structure form into the pruned data in the multi-tree structure form presented in text form.

[0119] Example 21. The apparatus of any one of Examples 14-20, wherein the data generation module comprises: an eighth determination module, configured to determine a plurality of fields in the data in the multi-tree structure form after being pruned and presented in text form, wherein the plurality of fields include a key field and a data field; and The first generating module is configured to replace the key field with a preset symbol to generate the normalized serialized data.

[0120] Example 22. According to the apparatus of any one of Examples 14-21, the language model is a first language model, wherein the attribution summary generation module comprises: a second generating module configured to generate an attribution summary with preset symbols based on the normalized serialized data by the first language model; and The third generating module is configured to normalize and deserialize the attribution summary with the preset symbol to generate the attribution summary for the data.

[0121] Example 23. The apparatus of any one of Examples 14-22, further comprising: a fourth generating module configured to generate training data for the first language model based on a predetermined format by the second language model, the training data including training node attribution data and training whole tree attribution data; and The training module is configured to train the first language model based on the training data.

[0122] Example 24. The apparatus of any one of Examples 14-23, wherein the fourth generation module comprises: The training label generation module is configured to generate a training label for the training data based on a preset prompt and the training data by the second language model.

[0123] Example 25. The apparatus of any one of Examples 14-24, further comprising: a ninth determination module, configured to determine whether the generated training data satisfies a legality check; and The removal module is configured to remove the generated training data that does not satisfy the legality check in response to the generated training data not satisfying the legality check.

[0124] Example 26. The apparatus of any one of Examples 14-25, further comprising: A first display module configured to display an attribution summary of node attribution data about the data in response to detecting a user interaction with the first control; and The second display module is configured to display an attribution summary of the entire tree of attribution data about the data in response to detecting a user interaction with the first control.

[0125] Example 27. An electronic device comprising: Processor; and A memory coupled to the processor, the memory having instructions stored therein, wherein when the instructions are executed by the processor, the electronic device performs actions, the actions comprising: Obtaining data in a text description format, the data including changes in the data within a date range; Based on the data in the text description form and the preset dimensions, an analysis tree is generated, the analysis tree including at least one of node attribution data and whole tree attribution data, the node attribution data indicating that the data is presented in a node hierarchical nested form, each node in the node attribution data including a key numerical description of the node, and the whole tree attribution data is determined based on a node disassembly path; Based on a first parameter and a second parameter, pruning the analysis tree to obtain pruned data in a multi-branch tree structure, wherein the first parameter indicates whether to prune the node based on the contribution of the node in the multi-branch tree structure, and the second parameter indicates whether to prune the node based on the importance of the node in the multi-branch tree structure; Normalizing and serializing the pruned data in the multi-tree structure form to generate normalized serialized data, wherein the normalized serialized data is presented in a text description form; and An attribution summary for the data is generated by a language model based on the normalized serialized data.

[0126] Example 28. The electronic device according to Example 27, wherein based on the first parameter and the second parameter, pruning the analysis tree to obtain pruned data in a multi-branch tree structure comprises: Converting the analysis tree into data in a multi-branch tree structure, wherein the multi-branch tree structure includes a plurality of nodes; and Based on the first parameter and the second parameter, pruning nodes in the multi-tree structure to obtain pruned data in the multi-tree structure form; and The pruned data in the multi-tree structure form is converted into the pruned data in the multi-tree structure form presented in text form.

[0127] Example 29. According to any one of Examples 27-28, the contribution is determined based on a contribution rate of a node, wherein based on the first parameter and the second parameter, pruning the nodes in the multi-branch tree structure to obtain pruned data in the form of the multi-branch tree structure comprises: In response to the analysis tree being node attribution data, traversing downward in sequence a plurality of nodes at each level from the level where the highest node of the data in the multi-branch tree structure corresponding to the node attribution data is located; Determine a maximum contribution rate, the maximum contribution rate indicating the contribution rate of a node having the largest contribution rate to the highest node among each node at the same level in the data in the multi-branch tree structure form corresponding to the node attribution data; Determine the ratio of the contribution rate of each node in each node of the same level in the data in the form of a multi-branch tree structure corresponding to the node attribution data to the maximum contribution rate; and In response to the ratio being less than the first parameter, in each node at the same level, each node whose ratio is less than the first parameter is pruned.

[0128] Example 30. The apparatus of any one of Examples 27-29, further comprising: In response to the analysis tree being adjusted based on the first parameter, sorting each node in the same level in the analysis tree; In response to the sorted analysis tree being node attribution data, determining the number of levels of each leaf node in the data in the multi-branch tree structure form corresponding to the node attribution data; and In response to the number of levels being greater than the second parameter, each node whose number of levels is greater than the second parameter and a previous non-leaf node of the node are pruned in a loop.

[0129] Example 31. The device according to any one of Examples 27-30, wherein the first parameter includes a disassembly depth and a candidate number of a disassembly path, the disassembly depth and the candidate number are related to the disassembly path, wherein based on the first parameter and the second parameter, pruning the nodes in the multi-tree structure to obtain the pruned data in the multi-tree structure form includes: In response to the analysis tree being the whole tree attributed data, traversing each leaf node in the data in the multi-branch tree structure form corresponding to the whole tree attributed data; Determine the contribution rate of each leaf node to the root node in the data in the form of a multi-branch tree structure corresponding to the whole tree attribution data; and Based on the disassembly depth, the number of candidates, and the contribution rate of each leaf node to the root node in the multi-branch tree structure data corresponding to the whole tree attribution data, it is determined whether to prune each node.

[0130] Example 32. The apparatus of any one of Examples 27-31, further comprising: In response to the analysis tree being adjusted based on the first parameter, sorting each node in the same level in the analysis tree; Determine the number of levels of each leaf node in the sorted data in the multi-branch tree structure corresponding to the whole tree attribution data; and Based on the second parameter and the number of levels, it is determined in a loop whether to prune each leaf node in the data in the multi-branch tree structure form corresponding to the whole tree attribution data.

[0131] Example 33. The apparatus of any one of Examples 27-32, wherein converting the pruned data in the multi-tree structure form into the pruned data in the multi-tree structure form presented in text form comprises: Depth-first traversal of the pruned data in the multi-branch tree structure; and The pruned data in the multi-tree structure form is converted into the pruned data in the multi-tree structure form presented in text form.

[0132] Example 34. The apparatus according to any one of Examples 27-33, wherein normalizing and serializing the pruned data in the multitree structure form to generate normalized serialized data comprises: Determine a plurality of fields in the data in the multi-tree structure form after the trimming presented in text form, wherein the plurality of fields include a key field and a data field; and The key field is replaced with a preset symbol to generate normalized serialized data.

[0133] Example 35. According to the device of any one of Examples 27-34, the language model is a first language model, wherein generating, by the language model based on the normalized serialized data, an attribution summary for the data comprises: The first language model generates an attribution summary with preset symbols based on the normalized serialized data; and The attribution summary with the preset symbol is normalized and deserialized to generate an attribution summary for the data.

[0134] Example 36. The apparatus of any one of Examples 27-35, further comprising: The second language model generates training data for the first language model based on a predetermined format, wherein the training data includes training node attribution data and training whole tree attribution data; and The first language model is trained based on the training data.

[0135] Example 37. The apparatus of any one of Examples 27-36, wherein generating, by a second language model based on a predetermined format, training data for the first language model comprises: The second language model generates a training label for the training data based on a preset prompt and the training data.

[0136] Example 38. The apparatus of any one of Examples 27-37, further comprising: Determining whether the generated training data satisfies a legality check; and In response to the generated training data not satisfying the legality check, the generated training data not satisfying the legality check is removed.

[0137] Example 39. The apparatus of any one of Examples 27-38, further comprising: In response to detecting a user interaction with the first control, displaying an attribution summary of node attribution data regarding the data; and In response to detecting a user interaction with the first control, an attribution summary of the entire tree of attribution data for the data is displayed.

[0138] Example 40. A computer-readable storage medium having one or more computer instructions stored thereon, wherein the one or more computer instructions are executed by a processor to implement the method according to any one of Examples 1 to 13.

[0139] Example 41. A computer program product tangibly stored on a computer-readable medium and comprising computer-executable instructions that, when executed by a device, cause the device to perform the method according to any one of Examples 1 to 13.

[0140] Although the disclosure has been described in language specific to structural features and / or methodological logical actions, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or actions described above. Instead, the specific features and actions described above are merely example forms of implementing the claims.

Claims

1. A method for data attribution, comprising: Obtaining data in a text description format, the data including changes in the data within a date range; Based on the data in the text description form and the preset dimensions, an analysis tree is generated, the analysis tree including at least one of node attribution data and whole tree attribution data, the node attribution data indicating that the data is presented in a node hierarchical nested form, each node in the node attribution data including a key numerical description of the node, and the whole tree attribution data is determined based on a node disassembly path; Based on a first parameter and a second parameter, pruning the analysis tree to obtain pruned data in a multi-branch tree structure, wherein the first parameter indicates whether to prune the node based on the contribution of the node in the multi-branch tree structure, and the second parameter indicates whether to prune the node based on the importance of the node in the multi-branch tree structure; Normalizing and serializing the pruned data in the multi-tree structure form to generate normalized serialized data, wherein the normalized serialized data is presented in a text description form; and An attribution summary for the data is generated by a language model based on the normalized serialized data.

2. The method according to claim 1, wherein based on the first parameter and the second parameter, pruning the analysis tree to obtain pruned data in the form of a multi-branch tree structure comprises: Converting the analysis tree into data in a multi-branch tree structure, wherein the multi-branch tree structure includes a plurality of nodes; as well as Based on the first parameter and the second parameter, pruning nodes in the multi-branch tree structure to obtain pruned data in the multi-branch tree structure form; as well as The pruned data in the multi-tree structure form is converted into the pruned data in the multi-tree structure form presented in text form.

3. The method according to claim 2, wherein the contribution is determined based on a contribution rate of a node, wherein based on the first parameter and the second parameter, pruning the nodes in the polytree structure to obtain pruned data in the polytree structure form comprises: In response to the analysis tree being node attribution data, traversing downward in sequence a plurality of nodes at each level from the level where the highest node of the data in the multi-branch tree structure corresponding to the node attribution data is located; Determine a maximum contribution rate, the maximum contribution rate indicating the contribution rate of a node having the largest contribution rate to the highest node among each node at the same level in the data in the multi-branch tree structure form corresponding to the node attribution data; Determine the ratio of the contribution rate of each node in each node of the same level in the data in the form of a multi-branch tree structure corresponding to the node attribution data to the maximum contribution rate; as well as In response to the ratio being less than the first parameter, in each node at the same level, each node whose ratio is less than the first parameter is pruned.

4. The method according to claim 3, further comprising: In response to the analysis tree being adjusted based on the first parameter, sorting each node in the same level in the analysis tree; In response to the sorted analysis tree being node attribution data, determining the number of levels of each leaf node in the data in the multi-branch tree structure form corresponding to the node attribution data; as well as In response to the number of levels being greater than the second parameter, each node whose number of levels is greater than the second parameter and the previous non-leaf node of the node are pruned in a loop.

5. The method according to claim 2, wherein the first parameter comprises a disassembly depth and a candidate number of a disassembly path, the disassembly depth and the candidate number are related to the disassembly path, wherein based on the first parameter and the second parameter, pruning the nodes in the multi-tree structure to obtain the pruned data in the multi-tree structure form comprises: In response to the analysis tree being the whole tree attributed data, traversing each leaf node in the data in the multi-branch tree structure form corresponding to the whole tree attributed data; Determine the contribution rate of each leaf node to the root node in the data in the form of a multi-branch tree structure corresponding to the whole tree attribution data; as well as Based on the disassembly depth, the number of candidates, and the contribution rate of each leaf node to the root node in the multi-branch tree structure data corresponding to the whole tree attribution data, it is determined whether to prune each node.

6. The method according to claim 5, further comprising: In response to the analysis tree being adjusted based on the first parameter, sorting each node in the same level in the analysis tree; Determine the number of levels of each leaf node in the sorted data in the multi-branch tree structure corresponding to the whole tree attribution data; as well as Based on the second parameter and the number of levels, it is determined in a loop whether to prune each leaf node in the data in the multi-branch tree structure form corresponding to the whole tree attribution data.

7. The method according to claim 2, wherein converting the pruned data in the multi-tree structure form into the pruned data in the multi-tree structure form presented in text form comprises: Depth-first traversal of the pruned data in the multi-branch tree structure; as well as The pruned data in the multi-tree structure form is converted into the pruned data in the multi-tree structure form presented in text form.

8. The method according to claim 7, wherein normalizing and serializing the pruned data in the multi-tree structure form to generate normalized serialized data comprises: Determine a plurality of fields in the data in the multi-tree structure form after being pruned and presented in text form, wherein the plurality of fields include a key field and a data field; as well as The key field is replaced with a preset symbol to generate normalized serialized data.

9. The method according to claim 1, wherein the language model is a first language model, wherein generating, by the language model based on the normalized serialized data, an attribution summary for the data comprises: The first language model generates an attribution summary with preset symbols based on the normalized serialized data; as well as The attribution summary with the preset symbol is normalized and deserialized to generate an attribution summary for the data.

10. The method according to claim 9, further comprising: The second language model generates training data for the first language model based on a predetermined format, wherein the training data includes training node attribution data and training whole tree attribution data; as well as The first language model is trained based on the training data.

11. The method according to claim 10, wherein generating training data for the first language model by the second language model based on a predetermined format comprises: The second language model generates a training label for the training data based on a preset prompt and the training data.

12. The method according to claim 11, further comprising: Determining whether the generated training data satisfies a legality check; as well as In response to the generated training data not satisfying the legality check, the generated training data not satisfying the legality check is removed.

13. The method according to claim 1, further comprising: In response to detecting a user interaction with the first control, displaying an attribution summary of node attribution data regarding the data; as well as In response to detecting a user interaction with the first control, an attribution summary of the entire tree of attribution data for the data is displayed.

14. An apparatus for data attribution, comprising: A data acquisition module is configured to acquire data in a text description form, wherein the data includes changes in the data within a date range; A parse tree generation module is configured to generate a parse tree based on the data in the text description form and preset dimensions, the parse tree including at least one of node attribution data and whole tree attribution data, the node attribution data indicating that the data is presented in a node hierarchical nested form, each node in the node attribution data including a key numerical description of the node, and the whole tree attribution data being determined based on a node disassembly path; A parse tree pruning module is configured to prune the parse tree based on a first parameter and a second parameter to obtain pruned data in a multi-branch tree structure, wherein the first parameter indicates whether to prune the node based on the contribution of the node in the multi-branch tree structure, and the second parameter indicates whether to prune the node based on the importance of the node in the multi-branch tree structure; A data generation module is configured to normalize and serialize the pruned data in the multi-tree structure form to generate normalized serialized data, wherein the normalized serialized data is presented in a text description form; as well as The attribution summary generating module is configured to generate an attribution summary for the data based on the normalized serialized data by a language model.

15. An electronic device, comprising: processor; as well as A memory coupled to the processor, the memory having instructions stored therein, wherein when the instructions are executed by the processor, the electronic device executes the method according to any one of claims 1 to 13.

16. A computer program product tangibly stored on a non-transitory computer readable medium and comprising computer executable instructions for performing the method according to any one of claims 1 to 13.

Images