Data dashboard generation method, device, equipment and medium

Abstract

The present application discloses a method, apparatus, device and medium for generating a data dashboard. The method includes: obtaining raw data to be processed, and generating an indicator data table based on the dimensional features of the raw data, wherein the dimensional features include target dimensional features; based on an indicator splitting strategy, sequentially determining indicator data used as splitting nodes from the indicator data table, and marking the hierarchical positions corresponding to the indicator data, wherein the indicator data are characteristic values of the dimensional features; based on the target dimensional features, the indicator data and their corresponding hierarchical positions, generating the data dashboard, thereby improving the rationality of indicator division in the construction of the indicator system.

Description

Technical Field

[0001] The present application generally relates to the field of data processing, specifically to artificial intelligence data processing, and more particularly to a method, apparatus, device, and medium for generating a data dashboard. Background Art

[0002] As internet businesses and products develop, massive amounts of data continue to be generated. Properly utilizing and displaying this data can lead to increasingly scientific business and product measurement. However, due to the sheer volume of data, how to properly display the key insights within this vast amount of data poses a significant challenge to business and R&D personnel.

[0003] Data partitioning in related technologies relies heavily on the knowledge base of business personnel and / or R&D personnel, such as Figure 1 As shown, different indicator system structures may be divided for the same set of data, which makes it inconvenient for business personnel to effectively obtain data information from the indicator system. Summary of the Invention

[0004] In view of the above-mentioned defects or deficiencies in the prior art, it is desired to provide a data dashboard generation method, device, equipment and medium to improve the rationality of indicator division in the construction of the indicator system.

[0005] In a first aspect, an embodiment of the present application provides a method for generating a data dashboard, comprising the following steps:

[0006] Obtaining raw data to be processed, and generating an indicator data table based on dimensional features of the raw data, wherein the dimensional features include target dimensional features;

[0007] Based on the indicator splitting strategy, the indicator data used as the splitting node is determined from the indicator data table in sequence, and the hierarchical position corresponding to the indicator data is marked. The indicator data is the characteristic value of the dimension feature;

[0008] Generate a data dashboard based on target dimension characteristics, indicator data and its corresponding hierarchical position.

[0009] In a second aspect, an embodiment of the present application provides a data dashboard generation device, comprising:

[0010] An acquisition module, configured to acquire raw data to be processed and generate an indicator data table based on dimensional features of the raw data, wherein the dimensional features include target dimensional features;

[0011] A splitting module is used to determine the indicator data used as split nodes from the indicator data table in sequence based on the indicator splitting strategy, and mark the hierarchical position corresponding to the indicator data. The indicator data is the characteristic value of the dimension feature;

[0012] The combination module is used to generate a data dashboard based on target dimension features, indicator data and their corresponding hierarchical positions.

[0013] In some embodiments, the splitting strategy includes a hierarchical splitting strategy, a splitting module configured to:

[0014] Based on the indicator data table, calculate the information purity corresponding to all the data in the indicator data table;

[0015] For each dimensional feature, determine the quantitative purity corresponding to the dimensional feature;

[0016] Based on the changing trend of the quantitative purity corresponding to the hierarchical splitting strategy, the dimensional features used for splitting are determined in sequence from the indicator data table, and the first order in which the dimensional features are split is used as the hierarchical position.

[0017] In some embodiments, the impact degree corresponding to the hierarchical splitting strategy changes from large to small. The splitting module is used to:

[0018] Compare the quantization purity corresponding to each dimension feature and extract the maximum value of at least one current quantization purity;

[0019] Mark the dimensional feature corresponding to the maximum value in the quantization purity as the target dimensional feature currently used for splitting, and record the first order in which the target dimensional feature is split;

[0020] Repeat the above steps until each dimension feature is marked as the target dimension feature.

[0021] In some embodiments, the splitting strategy further includes an attribute splitting strategy, a splitting module, configured to:

[0022] Based on the indicator data table, identifying the attribute type of at least one indicator data corresponding to each dimension feature, where the attribute type of the indicator data is determined according to the degree of continuity of the indicator data;

[0023] Based on the attribute type of the indicator data, determine the attribute splitting strategy for splitting the dimension features;

[0024] Based on the attribute splitting strategy, the dimension features are attribute split to obtain the indicator data used as the splitting node and its corresponding second order, and the second order is used as the hierarchical position.

[0025] In some embodiments, the split module is configured to:

[0026] When the attribute type is discrete, the dimension features are split based on the number of indicator data;

[0027] When the attribute type is continuous, the dimension features are split based on the greedy algorithm.

[0028] In some embodiments, the split module is configured to:

[0029] When the attribute type is discrete, each indicator data corresponding to the dimension feature is used as a split node, and the second order of the indicator data is randomly obtained;

[0030] When the attribute type is continuous, at least one indicator data for splitting is determined from the indicator data based on a greedy algorithm, and the second order of each split indicator data is recorded.

[0031] In some embodiments, the hierarchical position includes a first order in which the dimension features to which the indicator data belongs are split and a second order in which the indicator data is split between layers, and the combination module is configured to:

[0032] Arrange the indicator data vertically from the target dimension feature in the order of increasing first order, and arrange at least one indicator data belonging to the same dimension feature horizontally in the order of second order;

[0033] A linear connection relationship is established between two adjacent first-order indicator data to obtain a data dashboard.

[0034] In a third aspect, an embodiment of the present application provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method described in the embodiment of the present application when executing the program.

[0035] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium on which a computer program is stored, which, when executed by a processor, implements the method described in the embodiment of the present application.

[0036] This application splits the indicator data by combining the decision tree idea to construct an indicator system, which greatly improves the rationality of indicator division in the construction of the indicator system and the readability of the data dashboard.

[0037] Additional aspects and advantages of the present invention will be set forth in part in the description which follows and, in part, will be obvious from the description which follows, or may be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Other features, objects and advantages of the present application will become more apparent upon reading the detailed description of non-limiting embodiments made with reference to the following drawings:

[0039] Figure 1 This is a schematic diagram of the indicator system constructed in the relevant technology;

[0040] Figure 2 A schematic diagram of the architecture of a data dashboard generation method provided in an embodiment of the present application;

[0041] Figure 3 A flowchart of a data dashboard generation method proposed in an embodiment of the present application;

[0042] Figure 4 A flowchart of another data dashboard generation method proposed in an embodiment of the present application;

[0043] Figure 5 This is a schematic diagram of the data dashboard structure after dimensional features are divided according to an embodiment of the present application;

[0044] Figure 6 A flowchart of another data dashboard generation method proposed in an embodiment of the present application;

[0045] Figure 7 A flowchart of another data dashboard generation method proposed in an embodiment of the present application;

[0046] Figure 8 This is a structural diagram of a data dashboard generation device proposed in an embodiment of the present application;

[0047] Figure 9 A schematic diagram of the structure of a computer system of an electronic device or server suitable for implementing an embodiment of the present application is shown. DETAILED DESCRIPTION

[0048] The present application will be further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are merely for the purpose of explaining the relevant invention and are not intended to limit the invention. It should also be noted that, for ease of description, only portions relevant to the invention are shown in the accompanying drawings.

[0049] It should be noted that, in the absence of conflict, the embodiments and features of the embodiments in this application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0050] Artificial Intelligence (AI) refers to the theories, methods, techniques, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, to perceive the environment, acquire knowledge, and use that knowledge to achieve optimal results. In other words, AI is a comprehensive technology within computer science that seeks to understand the essence of intelligence and produce new intelligent machines that can respond in a manner similar to human intelligence. AI also studies the design principles and implementation methods of various intelligent machines, enabling them to possess the capabilities of perception, reasoning, and decision-making.

[0051] Artificial intelligence (AI) technology is a comprehensive discipline encompassing a wide range of fields, encompassing both hardware and software technologies. Foundational AI technologies generally include sensors, specialized AI chips, cloud computing, distributed storage, big data processing, operating / interaction systems, and mechatronics. AI software technologies primarily encompass computer vision, speech processing, natural language processing, and machine learning / deep learning.

[0052] Computer vision (CV) is the science of making machines "see." Specifically, it refers to machine vision, where cameras and computers replace the human eye in identifying, detecting, and measuring objects. This involves further processing the images, transforming them into images more suitable for human observation or transmission to instruments. As a scientific discipline, computer vision studies related theories and technologies, attempting to build artificial intelligence systems capable of extracting information from images or multidimensional data. Computer vision technologies typically include image processing, image recognition, image semantic understanding, image retrieval, optical character recognition (OCR), video processing, video semantic understanding, video content / behavior recognition, three-dimensional object reconstruction, 3D technology, virtual reality, augmented reality, simultaneous localization and mapping, and common biometric recognition technologies such as facial recognition and fingerprint recognition.

[0053] Key technologies in speech technology include automatic speech recognition (ASR), text-to-speech (TTS), and voiceprint recognition. Enabling computers to hear, see, speak, and feel is the future direction of human-computer interaction, with speech becoming one of the most promising methods of human-computer interaction.

[0054] Natural language processing (NLP) is a key area of research in computer science and artificial intelligence. It studies the theories and methods that enable effective communication between humans and computers using natural language. Natural language processing (NLP) integrates linguistics, computer science, and mathematics. Therefore, research in this field involves natural language—the language we use in everyday life—and is closely linked to the study of linguistics. Natural language processing technologies typically include text processing, semantic understanding, machine translation, robotic question answering, and knowledge graphs.

[0055] Machine learning (ML) is a multidisciplinary field that encompasses probability theory, statistics, approximation theory, convex analysis, and algorithmic complexity theory. It specifically studies how computers can simulate or implement human learning behaviors to acquire new knowledge or skills and reorganize existing knowledge structures to continuously improve their performance. Machine learning is at the core of artificial intelligence and the fundamental way to make computers intelligent. Its applications span all areas of AI. Machine learning and deep learning typically include techniques such as artificial neural networks, belief networks, reinforcement learning, transfer learning, inductive learning, and self-learning.

[0056] Autonomous driving technology usually includes high-precision maps, environmental perception, behavioral decision-making, path planning, motion control and other technologies. Autonomous driving technology has broad application prospects.

[0057] With the research and advancement of artificial intelligence technology, artificial intelligence technology has been studied and applied in many fields, such as common smart homes, smart wearable devices, virtual assistants, smart speakers, smart marketing, unmanned driving, autonomous driving, drones, robots, smart medical care, smart customer service, etc. It is believed that with the development of technology, artificial intelligence technology will be applied in more fields and play an increasingly important role.

[0058] The solutions provided in the embodiments of this application involve technologies such as machine learning of artificial intelligence, which are specifically illustrated by the following embodiments.

[0059] In order to describe this application more clearly, the following is an explanation of relevant technical terms:

[0060] Data visualization refers to the process of presenting data from large datasets in the form of graphs and images, and using data analysis and development tools to uncover unknown information within them. Data visualization primarily aims to convey information clearly and effectively through graphical means.

[0061] Data dashboards: Data dashboards are a platform for data visualization, showcasing data through effective page layout and visual design. As a platform for data visualization, data dashboards can include one or more of textual data and charts. A well-designed data dashboard intuitively displays the information contained in the data and facilitates developers to find and identify issues.

[0062] Indicator: A specific type of metadata that is calculated using specific logic to quantify the degree of development of a physical entity, such as the number of registered users or the amount of user payments. It is the intersection of business and data.

[0063] Dimension: The angle from which to analyze and observe indicators, such as time, age, gender, channel, etc.

[0064] Indicator system: Starting from the business, through modular and systematic dimension breakdown, we ultimately form a series of organically unified, relatively independent indicators that reflect product patterns and operating conditions. The principles of the indicator system include comprehensiveness, scientificity, and hierarchy. Comprehensiveness means that the indicator system should fully reflect all dimensions and aspects of the business. Scientificity means that the indicator calculation method is scientific and reasonable. Hierarchy means that the established indicator system creates conditions for further factor analysis, and the hierarchical relationship of the indicator system will contain mutually inclusive items.

[0065] Decision tree: A decision analysis method that uses a decision tree to construct a decision tree based on the known probabilities of various situations to calculate the probability that the expected value of the net present value is greater than or equal to zero, evaluate project risks, and determine its feasibility.

[0066] Currently, most data dashboards are developed by developers based on the data display needs of business personnel and then displayed in the interface. Such dashboards rely heavily on the business personnel's personal knowledge and understanding of the business, resulting in highly customized dashboards. This poses a huge challenge to both business personnel and R&D personnel.

[0067] For example, the division of indicators is highly dependent on the knowledge structure and understanding of the business of business personnel and R&D personnel; it requires good cooperation between business personnel and R&D personnel to complete, which greatly increases the development cost of the data dashboard; the display method of the data dashboard is customized and developed by R&D personnel based on the needs of business personnel. The same set of data is developed differently based on different needs. When the needs need to be adjusted after development, R&D personnel still need to intervene.

[0068] The data dashboard generation method of the embodiment of the present application can be applied to Figure 2 Please refer to the application environment shown. Figure 2 , Figure 2 This is a schematic diagram of the architecture of the data dashboard generation method provided in an embodiment of the present application.

[0069] like Figure 2 As shown, the user terminal 1 can communicate with the server 3 through the network 2. The user terminal 1 may include but is not limited to a personal computer, a laptop, a smart phone, a tablet computer, and a portable wearable device.

[0070] The user terminal 1 and the server 3 are directly or indirectly connected via an active or wireless network 2. Optionally, the wireless communication network or wired communication network may utilize standard communication technologies and / or protocols. The network is typically the Internet, but may also be any network, including but not limited to a local area network (LAN), a metropolitan area network (MAN), a wide area network (WAN), a mobile, wired or wireless network, a private network, or any combination of a virtual private network.

[0071] Server 3 can be an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms. Server 3 provides the generated data dashboard to user terminal 1 for viewing by business personnel and / or R&D personnel.

[0072] Specifically, the user terminal 1 can be used to collect raw data provided by at least one user, and then send the collected raw data to the server 3 via a wired or wireless network 2. The server 3 executes the data dashboard generation method proposed in the embodiment of the present application, generates a data dashboard based on the raw data provided by the user terminal 1, and sends it to the user terminal 1 corresponding to the business personnel and / or R&D personnel, so that the business personnel and / or R&D personnel can view the data dashboard information, and then analyze the information in the data dashboard to achieve product optimization, user feedback, etc.

[0073] For ease of understanding and explanation, the following Figure 2 The figures describe in detail the data dashboard generation method, apparatus, device and medium provided in the embodiments of the present application.

[0074] Figure 3 The present invention provides a flowchart of a data dashboard generation method according to an embodiment of the present invention. It should be noted that the execution entity of the data dashboard generation method according to this embodiment is a data dashboard generation device, which can be implemented by software and / or hardware. The data dashboard generation device in this embodiment can be configured in an electronic device or in a server for controlling the electronic device, which communicates with the electronic device and controls it.

[0075] Among them, the electronic devices in this embodiment may include but are not limited to personal computers, platform computers, smart phones, smart speakers and other devices. This embodiment does not specifically limit the electronic devices.

[0076] Step 101: obtain the original data to be processed, and generate an indicator data table based on the dimensional characteristics of the original data.

[0077] Among them, the original data is the data used to generate the data dashboard. The original data may include at least one of the input data entered by the user through the user terminal, the collected data obtained by the application installed on the user terminal through detection and collection of object behavior, or the evaluation data generated by the application on the user evaluation.

[0078] For example, when a user registers an account in an application, they are usually required to provide information such as gender and age. The application obtains the raw data for displaying the data dashboard by receiving the input data entered by the user. Alternatively, when the user makes an online payment, the application collects the amount paid by the user to obtain the raw data for displaying the data dashboard.

[0079] In one or more embodiments, each piece of raw data corresponds to a dimension feature. For example, the dimension corresponding to the gender data entered by the user is "gender", and the corresponding indicator value is "male" or "female". The dimension corresponding to the educational level data entered by the user is "education level", and the corresponding indicator value can be "high school", "undergraduate", "master", "doctor", etc.

[0080] Optionally, the indicator data table consists of dimension features and indicators, where each column of the indicator data table belongs to a dimension, and each row under the dimension feature label is the indicator value corresponding to each user, and the name corresponding to the indicator value is indicator data. For example, Table 1

[0081] Table 1

[0082] Gender (discrete) Age (discrete) Educational background (discrete) Score (continuous) Progress (target) male 18 undergraduate 10 10000 male 25 master 20 20000 female 20 undergraduate 10 15000 female 50 PhD 30 30000 male 15 high school 20 0 female 58 high school 10 1000

[0083] In Table 1, the dimensional features include "gender," "age," "education," "score," and "income." This means that the indicator value for the first user's gender is "male," the indicator value for age is "18," the indicator value for education is "bachelor's degree," the indicator value for score is "10," and the indicator value for income is "10000." The indicator data for gender includes "male" and "female," and the indicator data for education includes "high school," "bachelor's degree," "master's degree," and "doctoral degree."

[0084] In one or more embodiments, the target dimension features are determined based on business requirements, and the business requirements may be provided by business personnel.

[0085] It should be understood that each dimension feature in the indicator data table has a correlation with the target dimension feature. That is to say, when constructing an indicator data table based on the original data, the target dimension feature can be obtained first, and then at least one related dimension feature related to the target dimension feature can be determined based on the target dimension feature. The label row of the indicator data table can be constructed based on the related dimension feature and the target dimension feature. Then, the indicator values belonging to the related dimension feature and the target dimension feature can be filtered from the original data, and the indicator values can be filled into the indicator data table in the form of one row for each user, thereby forming the final indicator data table. Alternatively, the indicator data of at least one dimension can be obtained in a targeted manner, and an indicator data table can be constructed in a manner with dimension features as columns and users as rows, and then one dimension feature can be selected from at least one dimension feature in the indicator data table as the target dimension feature, and the other dimension features can be selected as the related dimension features.

[0086] Step 102: Based on the indicator splitting strategy, indicator data used as splitting nodes are determined in sequence from the indicator data table, and the hierarchical position corresponding to the indicator data is marked. The indicator data is the characteristic value of the dimension feature.

[0087] Optionally, the indicator splitting strategy can be a decision tree algorithm, such as a top-down greedy algorithm, to ensure that the dimensional feature with the best classification effect can be selected as the splitting node at each stage. Preferably, ID3, C4.5, CART algorithm, etc. can be used.

[0088] In one or more embodiments, a decision tree algorithm can be used to split the indicator data table, sort the dimensional features with the best classification effect in a top-down order to obtain the hierarchical position, and split at least one indicator value corresponding to the dimensional feature in each layer to obtain the corresponding split node, until all indicator data are classified or all dimensional features are used.

[0089] Step 103: Generate a data dashboard based on the target dimension features, indicator data and their corresponding hierarchical positions.

[0090] In one or more embodiments, the target dimension feature can be used as the root node of the data dashboard, and then the data dashboard nodes are set in a divergent distribution according to the indicator data and its corresponding hierarchical position, thereby forming a tree-like data display structure as the data dashboard.

[0091] Optionally, the data dashboard may be generated by a server and sent to a user terminal for display, or may be directly generated and displayed by the user terminal itself, which is not limited in this application.

[0092] Therefore, this application splits the indicator data by combining the decision tree idea to construct an indicator system, which greatly improves the rationality of indicator division in the construction of the indicator system and the readability of the data dashboard.

[0093] Based on the data dashboard generation method of the embodiment of the present application, the indicator system (data dashboard) can be constructed directly relying on the characteristics of the data itself, without relying heavily on the knowledge reserves and business understanding of business personnel and R&D personnel, so that the ultimately generated data dashboard is more in line with the correlation characteristics of the data itself. Moreover, the data dashboard can be generated independently according to the indicator splitting strategy, without the need for business personnel and R&D personnel to communicate with each other, which greatly reduces the demand communication time and improves the data feedback speed.

[0094] Moreover, this application adopts a splitting strategy to build a data dashboard. When data anomalies occur, R&D personnel only need to check whether the splitting of each layer or between layers meets the splitting strategy, which effectively reduces the difficulty for R&D personnel to find and discover problems.

[0095] It should also be noted that the construction of an indicator system generally includes two main aspects: configuring indicator data and configuring the hierarchical structure of the indicator system. In one or more embodiments proposed in the embodiments of this application, the indicator data is a single indicator or the indicator value observed under different dimensional characteristics, and the correlation between the various indicators is the hierarchical structure of the indicator system.

[0096] It should be noted that the hierarchical relationship of the indicator system can be displayed differently according to the number of dimensional features. Based on the idea of decision tree construction, the hierarchical relationship of the indicator system can be automatically determined. The key to learning the decision tree is to find the final dimensional features from the indicator data. Generally speaking, as the division proceeds, we hope that the samples contained in the branch nodes of the decision tree belong to the same category as much as possible, that is, the nodes become more and more pure.

[0097] In one or more embodiments, the splitting strategy includes a hierarchical splitting strategy, such as Figure 4 and Figure 5 As shown, step 102, based on the indicator splitting strategy, sequentially determines the indicator data used as the splitting node from the indicator data table and marks the hierarchical position corresponding to the indicator data, including:

[0098] Step 1021: Based on the indicator data table, calculate the information purity corresponding to all the data in the indicator data table.

[0099] Optionally, the Gini coefficient, information entropy, and error rate may be used to determine the information purity corresponding to all data in the data table. Preferably, information entropy is used to determine information purity, wherein the smaller the value of information entropy, the higher the purity.

[0100] Step 1022: For each dimensional feature, determine the quantization purity corresponding to the dimensional feature.

[0101] Among them, the information purity corresponding to all the data in the indicator data table can be the sum of the quantitative purity corresponding to each dimensional feature. Therefore, the information purity corresponding to the indicator data table can be split to obtain the quantitative purity corresponding to each dimensional feature.

[0102] Step 1023 : Based on the changing trend of the quantitative purity corresponding to the hierarchical splitting strategy, the dimensional features to be used for splitting are determined in sequence from the indicator data table, and the first order in which the dimensional features are split is used as the hierarchical position.

[0103] Optionally, the change trend of the quantitative purity can be from large to small, wherein information gain, information gain ratio, or Gini Index methods can be used to achieve this. For example, when using information gain to determine the dimensional features for splitting, if the information gain value is larger, it means that the purity of the dimensional feature will be lost more, the correlation between the dimensional feature and the target dimensional feature is closer or the influence is greater, then the dimensional feature should be placed in the upper layer of the decision tree.

[0104] In one or more embodiments, the change trend of the quantitative purity corresponding to the hierarchical splitting strategy is from large to small. Based on the change trend of the quantitative purity corresponding to the hierarchical splitting strategy, the dimensional features used for splitting are determined in sequence from the indicator data, and the first order in which the dimensional features are split is used as the hierarchical position, including: comparing the quantitative purity corresponding to each dimensional feature, extracting the maximum value of at least one current quantitative purity, marking the dimensional feature corresponding to the maximum value in the quantitative purity as the target dimensional feature currently used for splitting, and recording the first order in which the target dimensional feature is split, and repeating the above steps until each dimensional feature is marked as the target dimensional feature.

[0105] That is to say, after determining the quantization purity corresponding to each dimensional feature, multiple quantization purities can be compared with each other, or a descending table of quantization purity can be established to obtain the maximum quantization purity, and the dimensional feature corresponding to the maximum quantization purity is used as the target dimensional feature currently used for splitting. For example, if the current split is the first time, the dimensional feature corresponding to the maximum quantization purity is the first dimensional feature to be split, and the first order of this dimensional feature after being split is 1. Then, the dimensional feature with a sequence of 1 is excluded, and the quantization purity corresponding to the remaining dimensional features that have not been split is compared. The maximum quantization purity is selected from the quantization purity of the dimensional features that have not been split, and the first order of the dimensional feature corresponding to the maximum quantization purity is marked as 2, and so on, until all dimensional features are marked with the first order of being split.

[0106] Therefore, this application can use the decision tree idea to arrange the relevant dimensional features in descending order according to the relationship between them and the target dimensional features when constructing the data dashboard, which greatly facilitates the efficiency of business personnel in obtaining information from the data dashboard. Business personnel no longer need to make incorrect division attempts based on their own understanding of the relevant dimensional features. Moreover, the data dashboard can assist business personnel in better specifying subsequent work plans, such as deepening the users corresponding to the dimensional features with high relevance, or adjusting the strategies for users corresponding to the dimensional features with low relevance.

[0107] In one or more embodiments, the splitting strategy also includes an attribute splitting strategy, such as Figure 6 As shown, based on the indicator splitting strategy, indicator data used as split nodes are determined in sequence from the indicator data table, including:

[0108] Step 1024: Based on the indicator data table, identify the type of at least one indicator data corresponding to each dimensional feature. The attribute type of the indicator data is determined according to the degree of continuity of the indicator data.

[0109] Among them, the degree of continuity of indicator data can be divided into discrete and continuous types. The discrete type means that the indicator data (indicator values) are discontinuous with each other, such as the indicator data "male" and "female" corresponding to the gender characteristics in Table 1. The continuous type means that the indicator data are continuous with each other, such as the indicator data "10", "20" and "30" corresponding to the score characteristics in Table 1. The value interval of the indicator data is 10 and is continuous.

[0110] Step 1025 : Determine an attribute splitting strategy for splitting the dimensional features based on the attribute type of the indicator data.

[0111] Optionally, when the attribute type is discrete, the dimension features are split based on the number of indicator data; when the attribute type is continuous, the dimension features are split based on a greedy algorithm.

[0112] It should be understood that discrete indicator data are independent of each other and cannot be classified, and deleting the attribute type of any indicator data will affect the efficiency of business personnel and / or R&D personnel in obtaining information from the data dashboard. Continuous indicator data is continuous, and there is correlation between indicator data connected to each other. Therefore, a greedy algorithm can be used to split the indicator data of dimensional features to divide the continuous data into data segments to assist in generating data dashboards, effectively reducing the number of nodes in the layer where the continuous indicator data is located, reducing the difficulty of business personnel reading the map, improving the efficiency of data acquisition, and improving the readability of the data dashboard.

[0113] Step 1026 , based on the attribute splitting strategy, perform attribute splitting on the dimensional features to obtain indicator data used as splitting nodes and their corresponding second order, and use the second order as the hierarchical position.

[0114] Optionally, when the attribute type is discrete, each indicator data corresponding to the dimension feature is used as a splitting node, and the second order of the indicator data is randomly obtained; when the attribute type is continuous, at least one indicator data for splitting is determined from the indicator data based on a greedy algorithm, and the second order of each split indicator data is recorded.

[0115] That is, when the attribute type of the indicator data is discrete, each indicator data corresponding to the dimension feature can be used as a splitting node. For example, the two indicator data corresponding to the gender feature, "male" and "female", can be used as splitting nodes. In this case, the second order of each pair of indicator data can be randomly obtained. That is, the second order corresponding to the indicator data "male" can be 1, and the second order corresponding to the indicator data "female" can be 2, or the second order corresponding to the indicator data "female" can be 1, and the second order corresponding to the indicator data "male" can be 2. It should be understood that the second order of the randomly obtained indicator data is continuous and non-repetitive.

[0116] When the attribute of the indicator data is continuous, at least one indicator data for splitting is determined from the indicator data based on a greedy algorithm, wherein the greedy algorithm (also known as the greedy algorithm) means that when solving a problem, the best choice is always made at the moment. That is to say, the greedy algorithm is used to select the indicator data that has the greatest impact on the current dimension feature from a plurality of continuous indicator values as a splitting node, and so on, a preset number of indicator data are selected as splitting nodes, or until all indicator data are used as splitting nodes. Then, the order in which each indicator data is selected by the greedy algorithm is used as the second order. It should be understood that the marking method of the second order is the same as the marking method of the first order, and this application will not go into details here.

[0117] In one or more embodiments, the hierarchical position includes the first order in which the dimensional features to which the indicator data belongs are split and the second order in which the indicator data are split between layers. Based on the target dimensional features, the indicator data and their corresponding hierarchical positions, a data dashboard is generated, including: arranging the indicator data vertically from the target dimensional features in order of gradually increasing first order, and arranging at least one indicator data belonging to the same dimensional feature horizontally in order of second order; establishing a linear connection relationship between two adjacent indicator data of the first order to obtain a data dashboard.

[0118] That is to say, the indicator data is used as the nodes of the decision tree, and the linear connections are used as the edges between the decision tree nodes, so as to convert the indicator data table into a data dashboard.

[0119] Optionally, the vertical arrangement can be from top to bottom or from bottom to top, and the horizontal arrangement can be from left to right or from right to left, which is not limited in this application. Figure 5 What is given is a schematic diagram of arranging dimensional features from top to bottom.

[0120] In order to more clearly understand the inventive concept of this application, Figure 7 As shown, the data dashboard generation method includes:

[0121] In step 201, the server obtains raw data and processes the raw data to obtain an indicator data table.

[0122] Step 202: The server calculates the information purity corresponding to the indicator data table.

[0123] In step 203 , the server determines the quantization purity of each dimensional feature that has not been split.

[0124] In step 204 , the server selects any dimensional feature from the dimensional features that have not been split.

[0125] In step 205, the server determines whether the purity reduction is maximized by splitting the selected dimensional features.

[0126] If yes, go to step 206 ; if no, go back to step 204 .

[0127] Step 206: The server selects the dimension feature for splitting.

[0128] Step 207: The server determines the type of indicator data corresponding to the dimensional feature.

[0129] If the indicator data is discrete, step 208 is executed; if the indicator data is continuous, step 209 is executed.

[0130] In step 208 , the server uses each indicator data corresponding to the dimension feature as a split node and randomly obtains the order of the indicator data.

[0131] In step 209 , the server determines the indicator data for splitting in sequence from the indicator data based on a greedy algorithm, and records the splitting order.

[0132] Step 210: The server determines whether there is any unsplit indicator data.

[0133] If yes, then go back to step 203; if no, then go to step 211.

[0134] In step 211 , the server constructs a data dashboard based on the splitting order from top to bottom and from left to right.

[0135] In step 212, the server sends the data dashboard to the user terminal for display.

[0136] In summary, this application splits the indicator data by combining the decision tree idea to construct an indicator system, which greatly improves the rationality of indicator division in the construction of the indicator system and the readability of the data dashboard.

[0137] It should be noted that although the operations of the present method are described in a particular order in the drawings, this does not require or imply that the operations must be performed in this particular order, or that all illustrated operations must be performed to achieve desirable results.

[0138] Figure 8 This is a structural diagram of a data dashboard generation device proposed in an embodiment of the present application.

[0139] like Figure 8 As shown, the data dashboard generating device 10 proposed in the embodiment of the present application includes:

[0140] An acquisition module 11 is configured to acquire raw data to be processed and generate an indicator data table based on dimensional features of the raw data, wherein the dimensional features include target dimensional features;

[0141] A splitting module 12 is used to determine the indicator data used as split nodes from the indicator data table in sequence based on the indicator splitting strategy, and mark the hierarchical position corresponding to the indicator data. The indicator data is the characteristic value of the dimension feature;

[0142] The combination module 13 is used to generate a data dashboard based on target dimension features, indicator data and their corresponding hierarchical positions.

[0143] In some embodiments, the splitting strategy includes a hierarchical splitting strategy, and the splitting module 12 is configured to:

[0144] Based on the indicator data table, calculate the information purity corresponding to all the data in the indicator data table;

[0145] For each dimensional feature, determine the quantitative purity corresponding to the dimensional feature;

[0146] Based on the changing trend of the quantitative purity corresponding to the hierarchical splitting strategy, the dimensional features used for splitting are determined in sequence from the indicator data table, and the first order in which the dimensional features are split is used as the hierarchical position.

[0147] In some embodiments, the influence degree corresponding to the hierarchical splitting strategy changes from large to small. The splitting module 12 is configured to:

[0148] Compare the quantization purity corresponding to each dimension feature and extract the maximum value of at least one current quantization purity;

[0149] Mark the dimensional feature corresponding to the maximum value in the quantization purity as the target dimensional feature currently used for splitting, and record the first order in which the target dimensional feature is split;

[0150] Repeat the above steps until each dimension feature is marked as the target dimension feature.

[0151] In some embodiments, the splitting strategy further includes an attribute splitting strategy, and the splitting module 12 is configured to:

[0152] Based on the indicator data table, identifying the attribute type of at least one indicator data corresponding to each dimension feature, where the attribute type of the indicator data is determined according to the degree of continuity of the indicator data;

[0153] Based on the attribute type of the indicator data, determine the attribute splitting strategy for splitting the dimension features;

[0154] Based on the attribute splitting strategy, the dimension features are attribute split to obtain the indicator data used as the splitting node and its corresponding second order, and the second order is used as the hierarchical position.

[0155] In some embodiments, the splitting module 12 is configured to:

[0156] When the attribute type is discrete, the dimension features are split based on the number of indicator data;

[0157] When the attribute type is continuous, the dimension features are split based on the greedy algorithm.

[0158] In some embodiments, the splitting module 12 is configured to:

[0159] When the attribute type is discrete, each indicator data corresponding to the dimension feature is used as a split node, and the second order of the indicator data is randomly obtained;

[0160] When the attribute type is continuous, at least one indicator data for splitting is determined from the indicator data based on a greedy algorithm, and the second order of each split indicator data is recorded.

[0161] In some embodiments, the hierarchical position includes a first order in which the dimension features to which the indicator data belongs are split and a second order in which the indicator data is split between layers. The combining module 13 is configured to:

[0162] Arrange the indicator data vertically from the target dimension feature in the order of increasing first order, and arrange at least one indicator data belonging to the same dimension feature horizontally in the order of second order;

[0163] A linear connection relationship is established between two adjacent first-order indicator data to obtain a data dashboard.

[0164] In summary, this application splits the indicator data by combining the decision tree idea to construct an indicator system, which greatly improves the rationality of indicator division in the construction of the indicator system and the readability of the data dashboard.

[0165] It should be understood that the units or modules described in the data dashboard generating device 10 are similar to those described in the reference Figure 3 The steps in the described method correspond to each other. Therefore, the operations and features described above for the method are also applicable to the data dashboard generation device 10 and the units contained therein, and will not be repeated here. The data dashboard generation device 10 can be pre-implemented in the browser or other security application of the electronic device, or loaded into the browser or its security application of the electronic device by downloading or other means. The corresponding units in the data dashboard generation device 10 can cooperate with the units in the electronic device to implement the solution of the embodiment of the present application.

[0166] The several modules or units mentioned in the detailed description above are not necessarily divided into one module or unit. In fact, according to the embodiments of the present disclosure, the features and functions of two or more modules or units described above can be embodied in one module or unit. Conversely, the features and functions of one module or unit described above can be further divided into multiple modules or units to be embodied.

[0167] Reference below Figure 9 , Figure 9 A schematic diagram of the structure of a computer system of an electronic device or server suitable for implementing the embodiments of the present application is shown.

[0168] like Figure 9 As shown, the computer system includes a central processing unit (CPU) 901, which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) 902 or the program loaded from the storage part 908 into the random access memory (RAM) 903. Various programs and data required for the operation instructions of the system are also stored in the RAM 903. The CPU 901, ROM 902 and RAM 903 are connected to each other via a bus 904. An input / output (I / O) interface 905 is also connected to the bus 904.

[0169] The following components are connected to the I / O interface 905: an input section 906 including a keyboard, a mouse, and the like; an output section 907 including a cathode ray tube (CRT), a liquid crystal display (LCD), and a speaker; a storage section 908 including a hard disk; and a communication section 909 including a network interface card such as a LAN card or a modem. The communication section 909 performs communication processing via a network such as the Internet. A drive 910 is also connected to the I / O interface 905 as needed. A removable medium 911, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, and the like, is installed in the drive 910 as needed, so that a computer program read therefrom can be installed into the storage section 908 as needed.

[0170] In particular, according to the embodiment of the present application, the above reference flow chart Figure 2 The described process can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network via the communication section 909, and / or installed from a removable medium 911. When the computer program is executed by the central processing unit (CPU) 901, the above-mentioned functions defined in the system of the present application are executed.

[0171] It should be noted that the computer-readable medium shown in this application can be a computer-readable signal medium or a computer-readable storage medium, or any combination of the two. The computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or device, or any combination of the above. More specific examples of computer-readable storage media can include, but are not limited to: an electrical connection with one or more wires, 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), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this application, a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, device, or device. In this application, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, which carries computer-readable program code. This propagated data signal can take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium that can transmit, propagate, or transfer a program for use by or in conjunction with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium may be transmitted using any suitable medium, including but not limited to wireless, wire, optical cable, RF, or any suitable combination thereof.

[0172] The flowcharts and block diagrams in the accompanying drawings illustrate the possible implementation architecture, functions and operating instructions of the systems, methods and computer program products according to various embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a module, program segment, or a part of code, and the aforementioned module, program segment, or a part of code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order than the order marked in the accompanying drawings. For example, the boxes represented by two connections can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of the boxes in the block diagram and / or flowchart, can be implemented using a dedicated hardware-based system that performs the specified function or operating instruction, or can be implemented using a combination of dedicated hardware and computer instructions.

[0173] The units or modules involved in the embodiments described in the present application can be implemented by software or by hardware. The described units or modules can also be set in a processor. For example, they can be described as: a processor includes an acquisition module, a splitting module and a combination module. Among them, the names of these units or modules do not constitute a limitation on the unit or module itself under certain circumstances. For example, the acquisition module can also be described as "acquiring the original data to be processed, and generating an indicator data table based on the dimensional features of the original data, wherein the dimensional features include target dimensional features".

[0174] As another aspect, the present application further provides a computer-readable storage medium, which may be included in the electronic device described in the above embodiments, or may exist independently and not be incorporated into the electronic device. The computer-readable storage medium stores one or more programs, which, when used by one or more processors, execute the data dashboard generation method described in the present application.

[0175] The above description is merely a preferred embodiment of the present application and an illustration of the technical principles employed. Those skilled in the art should understand that the scope of disclosure in this application is not limited to the technical solutions formed by a specific combination of the above-mentioned technical features, but also encompasses other technical solutions formed by any combination of the above-mentioned technical features or their equivalents without departing from the aforementioned disclosed concepts. For example, a technical solution formed by replacing the above-mentioned features with (but not limited to) technical features with similar functions disclosed in this application.

Claims

1. A data dashboard generation method, characterized in that: The following steps are involved: Acquire raw data to be processed, and generate an indicator data table based on dimensional features of the raw data, wherein the dimensional features include target dimensional features; Based on the indicator splitting strategy, indicator data used as splitting nodes are determined in sequence from the indicator data table, and the hierarchical positions corresponding to the indicator data are marked, where the indicator data are characteristic values of the dimensional features; Generate the data dashboard based on the target dimension feature, the indicator data and the corresponding hierarchical position; The indicator splitting strategy is a decision tree algorithm. Based on the indicator splitting strategy, indicator data used as splitting nodes are sequentially determined from the indicator data table, and the hierarchical position corresponding to the indicator data is marked. A decision tree algorithm is used to split the indicator data table, and the dimensional features with the best classification effect are sorted in a top-down order to obtain the hierarchical position. At least one indicator value corresponding to the dimensional feature is split in each layer to obtain corresponding split nodes, until all indicator data are classified or all dimensional features are used.

2. The method according to claim 1, characterized in that The splitting strategy includes a hierarchical splitting strategy. The indicator-based splitting strategy sequentially determines indicator data used as splitting nodes from the indicator data table and marks the hierarchical position corresponding to the indicator data, including: Based on the indicator data table, calculating the information purity corresponding to all data in the indicator data table; For each of the dimensional features, determining the quantization purity corresponding to the dimensional feature; Based on the changing trend of the quantitative purity corresponding to the hierarchical splitting strategy, the dimensional features for splitting are determined in sequence from the indicator data table, and the first order in which the dimensional features are split is used as the hierarchical position.

3. The method according to claim 2, characterized in that The change trend of the quantitative purity corresponding to the hierarchical splitting strategy is from large to small. Based on the change trend of the quantitative purity corresponding to the hierarchical splitting strategy, the dimensional features for splitting are sequentially determined from the indicator data table, and the first order in which the dimensional features are split is used as the hierarchical position, including: Comparing the quantization purities corresponding to each of the dimensional features, and extracting the maximum value of at least one of the current quantization purities; Mark the dimensional feature corresponding to the maximum value in the quantization purity as the target dimensional feature currently used for splitting, and record the first order in which the target dimensional feature is split; Repeat the above steps until each of the dimensional features is marked as the target dimensional feature.

4. The method according to claim 1, wherein The splitting strategy also includes an attribute splitting strategy. The indicator-based splitting strategy sequentially determines indicator data used as splitting nodes from the indicator data table, including: Based on the indicator data table, identifying an attribute type of at least one indicator data corresponding to each of the dimensional features, where the attribute type of the indicator data is determined according to the degree of continuity of the indicator data; Determining the attribute splitting strategy for splitting the dimensional feature based on the attribute type of the indicator data; Based on the attribute splitting strategy, attribute splitting is performed on the dimensional feature to obtain the indicator data used as the splitting node and its corresponding second order, and the second order is used as the hierarchical position.

5. The method according to claim 4, characterized in that The determining, based on the attribute type of the indicator data, the attribute splitting strategy for splitting the dimensional feature includes: When the attribute type is discrete, the dimension feature is split based on the number of the indicator data; When the attribute type is continuous, the dimension feature is split based on a greedy algorithm.

6. The method according to claim 5, characterized in that The attribute splitting of the dimensional features based on the attribute splitting strategy to obtain the indicator data used as the splitting node includes: When the attribute type is discrete, each indicator data corresponding to the dimensional feature is used as the splitting node, and the second order of the indicator data is randomly obtained; When the attribute type is continuous, at least one indicator data for splitting is determined from the indicator data based on the greedy algorithm, and the second order of each split indicator data is recorded.

7. The method according to claim 1, characterized in that The hierarchical position includes a first order in which the dimension feature to which the indicator data belongs is split and a second order in which the indicator data is split between layers. Generating a data dashboard based on the target dimension feature, the indicator data, and their corresponding hierarchical positions includes: Arrange the indicator data vertically from the target dimensional feature in the order of gradually increasing the first order, and arrange at least one indicator data belonging to the same dimensional feature horizontally according to the second order; A linear connection relationship is established between two adjacent indicator data of the first order to obtain the data dashboard.

8. A data dashboard generating device, characterized in that: include: An acquisition module, configured to acquire raw data to be processed and generate an indicator data table based on dimensional features of the raw data, wherein the dimensional features include target dimensional features; A splitting module, configured to determine, based on an indicator splitting strategy, indicator data used as splitting nodes from the indicator data table in sequence, and mark hierarchical positions corresponding to the indicator data, wherein the indicator data is a characteristic value of the dimensional feature; A combination module, configured to generate the data dashboard based on the target dimension feature, the indicator data and the corresponding hierarchical position; The indicator splitting strategy is a decision tree algorithm. Based on the indicator splitting strategy, indicator data used as splitting nodes are sequentially determined from the indicator data table, and the hierarchical position corresponding to the indicator data is marked. A decision tree algorithm is used to split the indicator data table, and the dimensional features with the best classification effect are sorted in a top-down order to obtain the hierarchical position. At least one indicator value corresponding to the dimensional feature is split in each layer to obtain corresponding split nodes, until all indicator data are classified or all dimensional features are used.

9. The device according to claim 8, characterized in that The splitting strategy includes a hierarchical splitting strategy, and the splitting module is used to: Based on the indicator data table, calculating the information purity corresponding to all data in the indicator data table; For each of the dimensional features, determining the quantization purity corresponding to the dimensional feature; Based on the changing trend of the quantitative purity corresponding to the hierarchical splitting strategy, the dimensional features for splitting are determined in sequence from the indicator data table, and the first order in which the dimensional features are split is used as the hierarchical position.

10. The device according to claim 9, characterized in that The change trend of the quantitative purity corresponding to the hierarchical splitting strategy is from large to small, and the splitting module is used to: Comparing the quantization purities corresponding to each of the dimensional features, and extracting the maximum value of at least one of the current quantization purities; Mark the dimensional feature corresponding to the maximum value in the quantization purity as the target dimensional feature currently used for splitting, and record the first order in which the target dimensional feature is split; Repeat the above steps until each of the dimensional features is marked as the target dimensional feature.

11. The device according to claim 8, characterized in that The splitting strategy also includes an attribute splitting strategy, and the splitting module is used to: Based on the indicator data table, identifying an attribute type of at least one indicator data corresponding to each of the dimensional features, where the attribute type of the indicator data is determined according to the degree of continuity of the indicator data; Determining the attribute splitting strategy for splitting the dimensional feature based on the attribute type of the indicator data; Based on the attribute splitting strategy, attribute splitting is performed on the dimensional feature to obtain the indicator data used as the splitting node and its corresponding second order, and the second order is used as the hierarchical position.

12. The device according to claim 11, characterized in that The splitting module is used to: When the attribute type is discrete, the dimension feature is split based on the number of the indicator data; When the attribute type is continuous, the dimension feature is split based on a greedy algorithm.

13. The device according to claim 12, characterized in that The splitting module is used to: When the attribute type is discrete, each indicator data corresponding to the dimensional feature is used as the splitting node, and the second order of the indicator data is randomly obtained; When the attribute type is continuous, at least one indicator data for splitting is determined from the indicator data based on the greedy algorithm, and the second order of each split indicator data is recorded.

14. The device according to claim 8, characterized in that The hierarchical position includes a first order in which the dimension feature to which the indicator data belongs is split and a second order in which the indicator data is split between layers, and the combination module is used to: Arrange the indicator data vertically from the target dimensional feature in the order of gradually increasing the first order, and arrange at least one indicator data belonging to the same dimensional feature horizontally according to the second order; A linear connection relationship is established between two adjacent indicator data of the first order to obtain the data dashboard.

15. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, it implements the data dashboard generation method according to any one of claims 1 to 7.

16. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the data dashboard generation method according to any one of claims 1 to 7 is implemented.

17. A computer program product, characterized in that The computer program product comprises a computer program carried on a computer-readable medium, wherein the computer program includes instructions for executing the data dashboard generation method according to any one of claims 1 to 7.

Images