CN115298660A - Information mapping method, apparatus, medium, and program product based on artificial intelligence - Google Patents
Information mapping method, apparatus, medium, and program product based on artificial intelligence Download PDFInfo
- Publication number
- CN115298660A CN115298660A CN202280002306.3A CN202280002306A CN115298660A CN 115298660 A CN115298660 A CN 115298660A CN 202280002306 A CN202280002306 A CN 202280002306A CN 115298660 A CN115298660 A CN 115298660A
- Authority
- CN
- China
- Prior art keywords
- information
- information content
- label
- image
- artificial intelligence
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/5866—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using information manually generated, e.g. tags, keywords, comments, manually generated location and time information
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/35—Clustering; Classification
- G06F16/353—Clustering; Classification into predefined classes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
- G06F40/295—Named entity recognition
Abstract
The application discloses an information mapping method, device, equipment, medium and program product based on artificial intelligence, the method comprises: acquiring information content; determining an information subject label of the information content based on a preset algorithm model; judging whether a target theme label matched with the information theme label exists in the first image library or not, wherein the image in the first image library is marked with the corresponding theme label; if the first image library has the target subject label matched with the information subject label, the image corresponding to the target subject label is used as a cover image of the information content.
Description
Technical Field
The present application relates to the field of computer technology, and more particularly, to a method, apparatus, device, medium, and program product for information mapping based on artificial intelligence.
Background
In financial information operation work, the work of configuring the cover drawings is often complicated, the operator needs to find pictures with appropriate copyright and quality to configure the cover drawings when facing mass information every day, the work is repeated, and the effect is not high.
The current information distribution cover map mode mainly comprises two random distribution cover maps and a manual distribution cover map. When the cover picture is randomly matched, some pictures without specific subjects are randomly used as the cover picture, and the main defects are that the picture subjects are not clear and do not seem to be related to the cover picture, a user feels that the pictures and texts are irrelevant, and the list reading effect is poor. When the cover drawings are manually configured, the pictures related to a single information theme are manually found and uploaded to the information cover drawings, and the main defects are that the work is complicated and the configuration of a large amount of information requires a lot of manpower.
Disclosure of Invention
The embodiment of the application provides an information matching method, device, equipment, medium and program product based on artificial intelligence, which can realize intelligent configuration of a cover page, reduce labor cost, enhance the joint degree of information content and matching, and improve the efficiency and effect of information matching.
In one aspect, an embodiment of the present application provides an information mapping method based on artificial intelligence, where the method includes:
acquiring information content;
determining an information subject label of the information content based on a preset algorithm model;
judging whether a target theme label matched with the information theme label exists in a first image library or not, wherein the image in the first image library is marked with a corresponding theme label;
if the first image library has a target theme label matched with the information theme label, taking an image corresponding to the target theme label as a cover image of the information content.
On the other hand, the embodiment of the present application provides an information mapping apparatus based on artificial intelligence, the apparatus includes:
an acquisition unit for acquiring information content;
the determining unit is used for determining the information subject label of the information content based on a preset algorithm model;
the judging unit is used for judging whether a target theme label matched with the information theme label exists in a first image library or not, wherein the image in the first image library is marked with a corresponding theme label;
and the processing unit is used for taking the picture corresponding to the target theme label as a cover picture of the information content if the target theme label matched with the information theme label exists in the first image library.
In another aspect, an embodiment of the present application provides a computer device, where the computer device includes a processor and a memory, where the memory stores a computer program, and the processor is configured to execute the artificial intelligence based information mapping method according to any one of the above embodiments by calling the computer program stored in the memory.
In another aspect, an embodiment of the present application provides a computer-readable storage medium, where a computer program is stored, where the computer program is suitable for being loaded by a processor to perform the artificial intelligence based information mapping method according to any of the above embodiments.
In another aspect, the present application provides a computer program product, which includes computer instructions, and when executed by a processor, the computer instructions implement the artificial intelligence based information mapping method according to any of the above embodiments.
The embodiment of the application acquires the information content; determining an information subject label of the information content based on a preset algorithm model; judging whether a target theme label matched with the information theme label exists in the first image library or not, wherein the image in the first image library is marked with the corresponding theme label; if the first image library has the target subject label matched with the information subject label, the image corresponding to the target subject label is used as a cover image of the information content. According to the embodiment of the application, the information subject label is identified through the algorithm based on the artificial intelligence, so that the accuracy of information positioning is improved for information of various contents; and the label retrieval matching is carried out in the first image library, and the image corresponding to the target subject label matched with the information subject label in the first image library is used as the cover image of the information content, so that the intelligent configuration of the cover image can be realized, the labor cost is reduced, the attaching degree of the information content and the matched image is enhanced, and the efficiency and the effect of information matching are improved.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
Fig. 1 is a schematic flowchart of an information mapping method based on artificial intelligence according to an embodiment of the present disclosure.
Fig. 2 is a schematic view of a first application scenario provided in the embodiment of the present application.
Fig. 3 is a schematic view of a second application scenario provided in the embodiment of the present application.
Fig. 4 is a schematic diagram of a third application scenario provided in the embodiment of the present application.
Fig. 5 is a schematic diagram of a fourth application scenario provided in the embodiment of the present application.
Fig. 6 is a schematic view of a fifth application scenario provided in the embodiment of the present application.
FIG. 7 is a flowchart illustrating an artificial intelligence based information mapping method according to an embodiment of the present application.
Fig. 8 is a schematic structural diagram of an information mapping apparatus based on artificial intelligence according to an embodiment of the present application.
Fig. 9 is a schematic structural diagram of a server according to an embodiment of the present application.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
The embodiment of the application provides an information mapping method and device based on artificial intelligence, terminal equipment and a storage medium. Specifically, the information mapping method based on artificial intelligence in the embodiment of the present application may be executed by a computer device, where the computer device may be a terminal or a server. The terminal can be a smart phone, a tablet computer, a notebook computer, a desktop computer, a smart television, a smart sound box, a wearable smart device, a smart vehicle-mounted terminal and other devices, and can further comprise a client, wherein the client can be a financial client, a browser client or an instant messaging client and the like. The server may be an independent physical server, a server cluster or a distributed system formed by a plurality of physical servers, or a cloud server providing basic cloud computing services such as a cloud service, a cloud database, cloud computing, a cloud function, cloud storage, a network service, cloud communication, middleware service, a domain name service, a security service, a content distribution network service, a big data and artificial intelligence platform, but is not limited thereto.
The following are detailed below. It should be noted that the description sequence of the following embodiments is not intended to limit the priority sequence of the embodiments.
Referring to fig. 1 to 7, fig. 1 is a schematic flowchart of an artificial intelligence based information mapping method according to an embodiment of the present application, fig. 2 to 6 are schematic diagrams of application scenarios according to the embodiment of the present application, and fig. 7 is a flowchart of an artificial intelligence based information mapping method according to an embodiment of the present application. The information mapping method based on artificial intelligence can be applied to a server. The method comprises the following steps:
In some embodiments, the obtaining information content includes:
crawling information content based on an information content source, and warehousing the information content into an information content library; or
Responding to a warehousing request aiming at the information content sent by the background management equipment, acquiring the information content, and warehousing the information content into an information content library.
The essential part of information operation is information mapping, and the acquisition mode of information content can include that a content source crawls to automatically check and store in a warehouse or manually creates a new warehouse.
For example, based on copyright cooperation with the information source website, the information content is crawled from the information content source based on a crawler tool. The crawler tool is a program or script that automatically crawls the world wide web according to certain rules. The crawler tool initiates a Request to a target site through an HTTP (hyper text transport protocol) library, namely, a Request is sent, the Request can contain information such as additional headers and the like, and the response of a server is waited; if the server can normally respond, a Response is obtained, the content of the Response is the content of the page to be obtained, and the type can be HTML, json character string, binary data (such as picture video) and the like; the obtained content can be HTML and can be analyzed by a regular expression and a webpage analysis library; the obtained content can also be Json, can be directly converted into Json object analysis, is generally binary data, and can be stored or further processed; the information content crawled by the crawler tool can be stored as texts, can be stored in a database, or can be stored in a file with a specific format.
For example, the information content may be manually created and warehoused, the information content is acquired in response to a warehousing request for the information content sent by the background management device, and the information content is warehoused in the information content warehouse.
In some embodiments, after the obtaining the information content, the method further comprises:
performing information auditing on the information content based on a preset auditing rule, wherein the preset auditing rule at least comprises sensitive word matching and filtering rule verification;
and if the information content does not hit the sensitive words and the information content does not hit the filtering rules, executing a step of determining the information subject label of the information content based on a preset algorithm model.
In some embodiments, the method further comprises:
and if the information content hits the sensitive words and/or the information content hits the filtering rules, generating first prompt information which is not approved, and sending the information content and the first prompt information to the background management equipment.
For example, taking the information content obtained by crawling the information content source as an example, the crawling of the information content source occupies a large part of the source of the information content, most of the information stored in the information source in this way is not matched, and then the information subject label is obtained according to the information content to automatically generate a matched graph.
The information contents crawled and stored by the content source are firstly stored in the information content library, and before step 120, information auditing processes such as sensitive word matching, filtering rule checking and the like are required to be performed on the information contents through the information auditing process.
The sensitive word stock and the filtering rule word stock are pre-selected word stocks, text matching is performed on fields such as titles, information sources and texts of the information contents when the information contents are put in the word stock, if the sensitive words or the filtering words are hit, the examination is automatically judged not to be passed, and the operator is required to perform manual examination again to determine whether to perform the step 120. If the sensitive word is not hit and the filter word is not hit, it is automatically determined that the audit is passed, and step 120 is further performed.
And 120, determining the information subject label of the information content based on a preset algorithm model.
In some embodiments, the determining the information subject label of the information content based on the preset algorithm model includes:
and acquiring an information title text corresponding to the information text data in the information content, and processing the information title text based on the preset algorithm model to determine the information subject label of the information content.
In some embodiments, the predetermined algorithm model includes a physical extraction model and a keyword extraction model, and the processing the information title text based on the predetermined algorithm model to determine the information subject label of the information content includes:
processing the information title text based on an entity extraction model to obtain an entity type label of the information content, wherein the entity extraction model is used for extracting company, industry and name information in the information title text;
processing the information title text based on a keyword extraction model to obtain a keyword label of the information content, wherein the keyword extraction model is used for extracting industry keyword information in the information title text;
and determining the information subject label of the information content according to the entity type label and the keyword label.
In some embodiments, the determining the information subject label of the information content according to the entity type label and the keyword label includes:
generating a tag list according to the entity type tag and the keyword tag;
and determining the first ranked tag in the tag list as the information subject tag of the information content.
In the embodiment of the application, an information subject label API interface can be provided on the side of the preset algorithm model, and a label list is finally returned through the entity extraction model (denoted as M1) and the keyword extraction model (denoted as M2) according to the information title text transmitted from the back end. And the rear end selects the first label in the list as the information subject label for displaying.
The entity extraction model and the keyword extraction model of the API interface share the same model structure from the model level, but different entity labels are designed in the data collection stage of the entity extraction model and the keyword extraction model, so that the constructed training data are different, and finally the capability of extracting different results is achieved.
For example, the entity extraction model mainly performs data annotation on three most important entity types of companies, industries, names and the like concerned in the financial field.
For example, the keyword extraction model mainly performs data annotation on common and important words in financial information, such as: distension, stock market, harbor thigh, meilian store, etc.
After collecting the corresponding labeled data, a model for identifying the named entity can be constructed by utilizing a natural language processing pre-training model Bert and a global pointer module. And (3) respectively fine-tuning the models by using the labeled data from different sources to finally obtain 2 result models, namely M1 and M2.
The Bert model is built by an Embedding (Embedding) layer and 12 transformation (Transformer) layers, 1.1 hundred million parameters are shared, and the model parameters are very huge. In this project, the Bert model is used as a text _ encoder to perform feature extraction on the input information title text. The input information title text firstly passes through a word splitter (Tokenizer) in Bert to obtain a label (tokens) sequence with length of L, the tokens sequence further converts the token text into word id according to the mapping relation in the vocab to obtain an input tensor (tensor) of [1, L ], then the input tensor is input into the Bert model, the Bert is used as an encoder, the tensor of [1, L ] passes through an Embedding layer to obtain tensor of [ L, D ] dimension, the tensor is marked as R (D =768 because of being the Bert model), then the tensor inputs a full pointer layer to output an [ n _ labels, L, L ], wherein n _ labels are the total category number of the entity, for example: for an entity extraction model M1 for extracting company, industry and person names simultaneously, n _ labels =3; only the entity extracting model M2 of the keyword is extracted, and n _ labels =1.
The full pointer layer is used for extracting rich semantic information of the entity by using the text _ encoder, and the head and the tail of the entity are indicated by one pointer matrix at one time, so that the position of the entity in the original text can be quickly positioned for direct extraction. In this embodiment, a simplified version of a Multi-Head Attention (Multi-Head Attention) module is used to implement this function. The Multi-Head Attention module performs matrix calculation by three matrixes Q (query), K (all keys) and V (value), and then performs Scaled Dot-Product Attention calculation. Here, the Q and K matrices (both matrices of [ D, D ]) and the tentor of [ L, D ] dimension obtained above (the tentor is denoted as R) are used directly to reduce the dimension of the input [ L, D ] into the feature space of [ L, D ] (D =64 is usually < < D), denoted as Q and K, and the correlation formula is as follows:
q＝R·Q；k＝R·K；
wherein S is α (i, j) denotes a pointer matrix of the alpha-th entity, which has a shape of [ L, L]When there are n _ labels classes, each entity class will be calculated to obtain such a pointer matrix, so that the output of the whole full pointer layer is [ n _ labels, L]Tentor of (1). It is noted that S α The rows indicate the entity head positions and the columns indicate the entity tail positions, so S α Although the matrix is square, only the upper triangle part has practical significance, and the output of the lower triangle is directly not considered.
The classification output layer is used for extracting entities from the output pointer square matrix, and values larger than 0 in [ n _ labels, L, L ] are regarded as activated entity heads and tails, so that the layer converts the logits output by the model into a 0/1 binary square matrix, the activated entity heads and tails are set to be 1, and the rest are 0.
After receiving the information title text, the API firstly transmits the text into the model M1, if the model does not return a result, namely the entity extraction model M1 does not extract any company, industry and name information, the API continuously transmits the title into the keyword extraction model M2, and further extracts keywords.
As shown in the application scenario diagram of fig. 2, in the training phase, the obtained information text data is manually labeled, for example, data labeling is performed on three most important entity types, such as company, industry, and name, in the information text data, so as to obtain first labeled data. And performing data annotation aiming at the industry keywords in the information text data to obtain second annotation data. Inputting the first labeled data into a preset algorithm model for model training, for example, pre-training a plurality of transformations layers, a global pointer layer and a classification output layer by using the first labeled data and a Bert model to obtain an entity extraction model M1). Inputting the second labeled data into a preset algorithm model for model training, for example, pre-training a plurality of transformations, a global pointer layer and a classification output layer by using the second labeled data and the Bert model to obtain a keyword extraction model M2).
In the application stage, inputting the information title text into the entity extraction model M1 for the 1 st time to extract company, industry and name information in the information title text and output a first result set; then, judging whether the first result set is empty or not, and if the first result set is not empty, classifying entity type labels contained in the first result set into a final result set; if the first result set is empty, the keyword extraction model M2 is input for the 2 nd time of the information title text to extract the industry keyword information in the information title text, a second result set is output, and the keyword labels contained in the second result set are classified into a final result set. In the final result set, a tag list can be constructed, and the tag ranked first in the list is selected as the information subject tag for display at the back end. That is, when the algorithm provides a plurality of tags with the same frequency, the information side can select the first tag for matching the image.
In the method, the entity extraction model M1 and the keyword extraction model M2 are obtained based on artificial intelligence training, and finally a label list is returned, so that the label with the first rank in the back-end selection list is displayed as the information subject label, and the accuracy of determining the information subject label is improved.
For example, after the information topic tag is generated through the preset algorithm model, searching and matching are performed in the first gallery to find whether a target topic tag matched with the information topic tag exists in the first gallery.
For example, as shown in the application scenario diagram shown in fig. 3, an information management platform interface is displayed on the background management device, and a first gallery (such as the tag gallery shown in fig. 3) may be maintained on the information management platform interface, where when each picture in the first gallery is maintained on the information management platform interface, information such as a thumbnail of the picture, ID information, warehousing time, a corresponding subject tag, hit frequency, and the like may be displayed.
After an information crawls from a content source to automatically check and store in a warehouse or manually creates a new warehouse, a cover page picture needs to be allocated to the information content. After information crawls into a library or is artificially created and stored, the intelligent matching of the cover page of the information content can be completed by extracting the information subject label of the information content according to a preset algorithm model and matching the pictures in the image library according to the information subject label. If the first image library has the target subject label matched with the information subject label, the image corresponding to the target subject label is used as a cover image of the information content.
In some embodiments, if a target topic tag matching the information topic tag exists in the first gallery, taking a picture corresponding to the target topic tag as a cover picture of the information content includes:
if a target subject label matched with the information subject label exists in the first image library and a plurality of images corresponding to the target subject label exist, selecting one image from the plurality of images corresponding to the target subject label to be used as a cover image of the information content.
For example, when a plurality of pictures corresponding to the target theme tag are matched based on the information theme tag, one picture can be randomly selected from the plurality of pictures corresponding to the target theme tag to serve as a cover page picture of the information content. For example, one picture with the largest hit frequency can be selected from the multiple pictures corresponding to the target theme tag according to the hit frequency to serve as a cover page of the information content, so that the most frequently used picture in the target theme tag is selected for matching, and the most popular image is as aesthetic as possible. For example, one of the plurality of pictures corresponding to the target topic tag with the smallest hit frequency can be selected as the cover page of the information content according to the hit frequency, so as to select the pictures which are rarely used in the target topic tag for matching, thereby increasing the freshness of matching pictures.
In some embodiments, the method further comprises:
if the target subject label matched with the information subject label does not exist in the first image library, a random image is obtained from a second image library to serve as a cover image of the information content, wherein the image in the first image library is not labeled with the subject label.
For example, as shown in the application scenario diagram of fig. 4, an information management platform interface is displayed on the background management device, and a second gallery (e.g., the general gallery shown in fig. 3) may be maintained on the information management platform interface, wherein when each picture in the second gallery is maintained on the information management platform interface, a thumbnail of the picture and ID information may be displayed.
For example, for a small number of information theme tags that are not maintained in the first gallery, abstract pictures without specific themes in the second gallery may be used as cover art pictures of the information content.
In some embodiments, the method further comprises:
when a cover page layout configuration request aiming at the information content and sent by a background management device is received, determining a cover page layout of the information content according to configuration information carried in the cover page layout configuration request.
For example, after the information is checked and stored, the operator can still re-edit the related information corresponding to the information content. For example, if the cover page generated by automatic matching does not meet the requirement yet, the information content can also be subjected to manual intervention matching, the information personnel can input a cover page configuration request aiming at the information content on the background management equipment, the configuration information of the cover page configuration request carries the specified cover page, and when the server receives the cover page configuration request of the information content sent by the background management equipment, the server determines the cover page of the information content according to the configuration information carried in the cover page configuration request, so that the flexibility of matching is greatly increased, and the operation efficiency is improved.
In some embodiments, the method further comprises:
when a cover page display request aiming at the information content sent by a client is received, sending the cover page of the information content to the client so as to display the cover page of the information content on an information preview interface of the client.
In some embodiments, the method further comprises:
when receiving a trigger operation aiming at the cover page picture sent by the client, sending an information stream of the information content to the client so as to display the information stream of the information content on an information preview interface of the client, wherein the information stream comprises at least one of information text data and information multimedia data corresponding to the information content.
For example, when the user is interested in the information content corresponding to the cover page, the server can push the information stream of the information content to the client by triggering operations such as clicking the cover page, pressing the cover page for a long time, and the like. The information stream may include information text data, information multimedia data, and the like, wherein the information multimedia data may include multimedia resources such as images, animations, audio, video, and the like. The information text data may include information title, ID, distribution time, information content details, information content distribution end information, and the like.
In some embodiments, the method further comprises:
acquiring matching record data in a preset time period, wherein each matching record comprises paired information subject labels and an artificial intelligence-based information matching mode corresponding to the information subject labels;
counting the occurrence frequency of each information subject label in the preset time period according to the matching image recording data to obtain label statistical data;
according to the image matching record data, counting the data proportion of each artificial intelligence-based information image matching mode in the preset time period to obtain image matching statistical data;
and sending the tag statistical data and the map matching statistical data to a background management device so as to display the tag statistical data and the map matching statistical data on the background management device.
In order to continuously enrich the map library of the label map matching and optimize the accuracy of map matching, statistical label statistics and map matching statistics need to be added, and operators can continuously maintain the map library according to statistical results to improve the accuracy of intelligent map matching. And the tag statistical data and the mapping statistical data of the information content are independently displayed in the background, so that the problem of disk duplication is conveniently positioned.
For example, each mapping record is recorded, and the label statistics and mapping statistics are used as the data source statistics. The data statistics is carried out based on the matching chart record data recorded in the preset time period, so that the tag statistical data and the matching chart statistical data are determined. For example, the preset period may be one of the last week, last month, last three months, last year, etc. periods.
For the tag statistical data, because the data volume of the data recorded by matching the graph is large, the data volume generated every day may reach tens of thousands, and the performance requirements cannot be met only by means of aggregation statistics of a database, the daily tag statistical data needs to be counted according to the day, for example, as shown in an application scene graph shown in fig. 5, when an operator inquires the tag statistical data, a preset time period, such as the latest month, may be input through an information management platform interface displayed on the background management device, and the preset time period is sent to the server, the server inquires the single-day tag statistical data within the preset time period range according to the preset time period, aggregates the single-day tag statistical data into the tag statistical data within the preset time period, and returns the tag statistical data within the preset time period to the background management device for display. By displaying the tag statistical data on the background management equipment, an operator can make targeted statistics on tags appearing at high frequency when maintaining a first gallery (such as the tag gallery shown in fig. 3), and information content of high-frequency information topics is covered as much as possible, so that gallery operation is more efficient. For example, the operator finds that the oil price is a label with high frequency, that is, maintains a plurality of related cover drawings for the oil price label, so that the intelligent matching drawing will be effective when the information content of the same label appears subsequently. As shown in fig. 5, when the tags are counted, the occurrence frequency of the information subject tags is counted at the same time, so as to facilitate the targeted maintenance of the pictures.
For the chart matching statistical data, because the data volume of the chart matching recorded data is large, the required statistical result cannot be obtained by simply depending on aggregation statistics of the database, the chart matching statistical data within 30 days before the statistics of each day are required to be stored in the database, and the corresponding chart matching statistical data can be inquired from the database according to the inquiry date and returned when the chart matching statistical data are checked. As shown in the application scenario diagram of fig. 6, when an operator queries the statistical data of map matching, the operator may input a preset time period, for example, push forward from 2022-5-17 for 7 days, through an interface of an information management platform displayed on the background management device, and send the preset time period to the server, and the server queries the statistical data of single-day map matching within the preset time period according to the preset time period, aggregates the statistical data of map matching within the preset time period, and returns the statistical data of map matching within the preset time period to the background management device for display. The on-line actual data ratio (coverage rate) of various matching types can be displayed through the statistical data of matching, so that the intelligent matching mechanism can be objectively evaluated, and the coverage effect of the statistical gallery can be truly quantized. For example, the matching types may include label matching, random matching and manual matching, and in the statistical data of matching shown in fig. 6, the coverage rate of label matching is 3.40%, the coverage rate of random matching is 96.40%, and the coverage rate of manual matching is 0.20%.
In some embodiments, the method further comprises:
when an update request aiming at the first gallery sent by the background management equipment is received, at least one of the pictures and the theme labels in the first gallery is updated according to the update request.
For example, the first gallery may be maintained manually, and the related cover art is maintained for the common information subject label. An operator can maintain the first gallery through an information management platform interface displayed by the background management equipment, input an updating request aiming at the first gallery through the information management platform interface and send the updating request to the server, so that the server updates at least one of the pictures and the theme labels in the first gallery according to the updating request. For example, if the update request is to update the first theme tag corresponding to the first picture, the first theme tag corresponding to the first picture in the first gallery is updated according to the update request. For example, if the update request is to update the second picture corresponding to the second theme tag, the second picture corresponding to the second theme tag in the first gallery is updated according to the update request. For example, if the update request is to add a third picture and a third theme tag corresponding to the third picture, the third picture and the third theme tag corresponding to the third picture are added to the first gallery according to the update request.
In some embodiments, the method further comprises:
and if detecting that the same theme label as the label to be updated carried in the updating request exists in the first gallery, generating second prompt information representing repeated labels, and refusing to update the first gallery.
For example, a plurality of same labels cannot be created in one picture, if it is detected that the same theme label as the label to be updated carried in the update request already exists in the first gallery, second prompt information representing that the labels are repeated is generated, the first gallery is refused to be updated, and the second prompt information is sent to the background management device so as to display the second prompt information on the background management device to prompt an operator that the same label already exists.
Referring to fig. 7, for better explaining the information mapping method based on artificial intelligence provided in the embodiment of the present application, a flow of the information mapping method based on artificial intelligence provided in the embodiment of the present application can be summarized as the following steps:
s1, a server acquires information content;
s1.1, the server crawls information contents based on an information content source and stores the information contents into an information content library;
s1.2, the server responds to a storage request aiming at the information content sent by the background management equipment, acquires the information content and stores the information content into an information content library;
s1.3, the server performs information audit on the information content based on preset audit rules, wherein the preset audit rules at least comprise sensitive word matching and filtering rule verification;
s1.4, if the information content hits the sensitive words and/or the information content hits the filtering rules, the server determines that the information examination is not passed, generates first prompt information that the examination is not passed, and sends the information content and the first prompt information to the background management equipment;
s1.5, displaying first prompt information by the background management equipment;
s1.6, if the sensitive words are not hit by the information content and the filtering rule is not hit by the information content, the server determines that the information is approved and further executes the step S2;
s2, the server determines an information subject label of the information content based on a preset algorithm model;
s2.1, the server processes the information title text of the information content based on an entity extraction model to obtain an entity type label of the information content, wherein the entity extraction model is used for extracting company, industry and name information in the information title text;
s2.2, the server processes the information title text of the information content on the basis of a keyword extraction model to obtain a keyword label of the information content, wherein the keyword extraction model is used for extracting industry keyword information in the information title text;
s2.3, the server determines an information subject label of the information content according to the entity type label and the keyword label;
s3, the server judges whether a target theme label matched with the information theme label exists in the first image library or not, and the images in the first image library are marked with corresponding theme labels;
s4, if the target theme label matched with the information theme label exists in the first image library, the server takes the image corresponding to the target theme label as a cover image of the information content;
s5, if the target theme tag matched with the information theme tag does not exist in the first image library, the server acquires a random image from the second image library as a cover image of the information content, and the image in the first image library is not labeled with the theme tag;
s6, when the server receives a cover page configuration request aiming at the information content and sent by the background management equipment, determining a cover page of the information content according to configuration information carried in the cover page configuration request;
s7, when the server receives a cover page image display request aiming at the information content sent by the client, sending the cover page image of the information content to the client;
s8, displaying a cover page image of the information content on an information preview interface of the client;
s9, when the server receives a trigger operation aiming at the cover page image sent by the client, sending an information stream of the information content to the client;
s10, displaying an information flow of the information content on an information preview interface of the client, wherein the information flow comprises at least one of information text data and information multimedia data corresponding to the information content;
s11, the server acquires image matching record data in a preset time period, wherein each image matching record comprises paired information subject labels and a determination mode of a cover image corresponding to the information subject labels;
s12, the server counts the occurrence frequency of each information subject label in a preset time period according to the image matching record data to obtain label statistical data;
s13, the server counts the data proportion of the determined mode of each cover map within a preset time period according to the map matching record data to obtain map matching statistical data;
s14, the server sends the tag statistical data and the map matching statistical data to background management equipment;
s15, displaying the tag statistical data and the map matching statistical data by the background management equipment;
s16, when the server receives an updating request aiming at the first gallery and sent by the background management equipment, updating at least one of the pictures and the subject labels in the first gallery according to the updating request;
s17, if the fact that the theme label identical to the label to be updated carried in the updating request exists in the first image library is detected, the server generates second prompt information representing repeated labels and refuses to update the first image library; in step S17, the second prompt message may also be sent to the backend management device.
And S18, displaying the second prompt message by the background management equipment.
All the above technical solutions can be combined arbitrarily to form the optional embodiments of the present application, and are not described herein again.
The information matching method based on artificial intelligence provided by the embodiment of the application obtains information content; determining an information subject label of the information content based on a preset algorithm model; judging whether a target theme label matched with the information theme label exists in the first image library or not, wherein the image in the first image library is marked with the corresponding theme label; if the first image library has the target subject label matched with the information subject label, the image corresponding to the target subject label is used as a cover image of the information content. According to the embodiment of the application, the information subject labels are identified through the algorithm, the labels are retrieved and matched in the first image library, the image corresponding to the target subject label matched with the information subject labels in the first image library is used as the cover image of the information content, the cover image can be configured intelligently, the labor cost is reduced, the attaching degree of the information content and the matched image is enhanced, and the efficiency and the effect of information matching are improved.
In order to better implement the artificial intelligence-based information mapping method according to the embodiment of the present application, the embodiment of the present application further provides a client. Referring to fig. 8, fig. 8 is a schematic structural diagram of an information mapping apparatus based on artificial intelligence according to an embodiment of the present application. The artificial intelligence-based information mapping apparatus 200 may include:
an obtaining unit 210, configured to obtain information content;
a determining unit 220, configured to determine an information subject label of the information content based on a preset algorithm model;
a determining unit 230, configured to determine whether a target theme tag matching the information theme tag exists in a first gallery, where a picture in the first gallery is labeled with a corresponding theme tag;
the processing unit 240 is configured to, if a target theme tag matching the information theme tag exists in the first gallery, use a picture corresponding to the target theme tag as a cover picture of the information content.
In some embodiments, the determining unit 220 is specifically configured to: and acquiring an information title text corresponding to information text data in the information content, and processing the information title text based on the preset algorithm model to determine an information subject label of the information content.
In some embodiments, the predetermined algorithm model includes an entity extraction model and a keyword extraction model, and the determining unit 220 is specifically configured to, when processing the information title text based on the predetermined algorithm model to determine the information subject label of the information content: processing the information title text based on an entity extraction model to obtain an entity type label of the information content, wherein the entity extraction model is used for extracting company, industry and name information in the information title text; processing the information title text based on a keyword extraction model to obtain a keyword label of the information content, wherein the keyword extraction model is used for extracting industry keyword information in the information title text; and determining the information subject label of the information content according to the entity type label and the keyword label.
In some embodiments, when determining the information subject label of the information content according to the entity type label and the keyword label, the determining unit 220 is specifically configured to: generating a tag list according to the entity type tag and the keyword tag; and determining the first ranked tag in the tag list as the information subject tag of the information content.
In some embodiments, the obtaining unit 210 is specifically configured to: crawling information content based on an information content source, and warehousing the information content into an information content library; or responding to a warehousing request aiming at the information content sent by the background management equipment, acquiring the information content, and warehousing the information content into an information content library.
In some embodiments, the artificial intelligence based information mapping apparatus 200, after being configured to obtain the information content, is further configured to: performing information auditing on the information content based on a preset auditing rule, wherein the preset auditing rule at least comprises sensitive word matching and filtering rule verification; and if the information content does not hit the sensitive words and the information content does not hit the filtering rules, executing a step of determining the information subject label of the information content based on a preset algorithm model.
In some embodiments, the artificial intelligence based information mapping apparatus 200 is further configured to: and if the information content hits the sensitive words and/or the information content hits the filtering rules, generating first prompt information which is not approved, and sending the information content and the first prompt information to the background management equipment.
In some embodiments, the processing unit 240 is further configured to: if the target theme tag matched with the information theme tag does not exist in the first image library, acquiring a random image from a second image library as a cover image of the information content, wherein the image in the first image library is not labeled with the theme tag.
In some embodiments, the processing unit 240 is further configured to: when a cover page layout configuration request aiming at the information content and sent by a background management device is received, determining a cover page layout of the information content according to configuration information carried in the cover page layout configuration request.
In some embodiments, the processing unit 240 is further configured to: when a cover page image display request aiming at the information content sent by a client is received, sending the cover page image of the information content to the client so as to display the cover page image of the information content on an information preview interface of the client.
In some embodiments, the processing unit 240 is further configured to: when receiving a trigger operation aiming at the cover page picture sent by the client, sending an information stream of the information content to the client so as to display the information stream of the information content on an information preview interface of the client, wherein the information stream comprises at least one of information text data and information multimedia data corresponding to the information content.
In some embodiments, the processing unit 240 is further configured to: acquiring matching record data in a preset time period, wherein each matching record comprises paired information subject labels and a determination mode of a cover map corresponding to the information subject labels; counting the occurrence frequency of each information subject label in the preset time period according to the matching image recording data to obtain label statistical data; according to the image matching record data, counting the data proportion of the determined mode of each cover image in the preset time period to obtain image matching statistical data; and sending the tag statistical data and the map matching statistical data to a background management device so as to display the tag statistical data and the map matching statistical data on the background management device.
In some embodiments, the processing unit 240 is further configured to: when an update request aiming at the first gallery sent by the background management equipment is received, at least one of the pictures and the theme labels in the first gallery is updated according to the update request.
In some embodiments, the processing unit 240 is further configured to: and if detecting that the same theme label as the label to be updated carried in the updating request exists in the first gallery, generating second prompt information representing repeated labels, and refusing to update the first gallery.
In some embodiments, the processing unit 240 is configured to: if a target subject label matched with the information subject label exists in the first image library and a plurality of images corresponding to the target subject label exist, selecting one image from the plurality of images corresponding to the target subject label to be used as a cover image of the information content.
All the above technical solutions can be combined arbitrarily to form the optional embodiments of the present application, and are not described herein again.
It should be understood that the embodiments of the information mapping apparatus and the embodiments of the method based on artificial intelligence can correspond to each other, and similar descriptions can be made with reference to the embodiments of the method. To avoid repetition, further description is omitted here. Specifically, the information mapping apparatus based on artificial intelligence shown in fig. 8 can execute the above-mentioned information mapping method embodiment based on artificial intelligence, and the above-mentioned and other operations and/or functions of each unit in the information mapping apparatus based on artificial intelligence respectively implement the corresponding processes of the above-mentioned method embodiment, and for brevity, are not described again here.
Optionally, the present application further provides a computer device, which includes a memory and a processor, where the memory stores a computer program, and the processor implements the steps in the foregoing method embodiments when executing the computer program.
Fig. 9 is a schematic structural diagram of a computer device according to an embodiment of the present application, where the computer device may be a terminal or a server. As shown in fig. 9, the computer apparatus 300 may include: a communication interface 301, a memory 302, a processor 303 and a communication bus 304. The communication interface 301, the memory 302 and the processor 303 realize mutual communication through a communication bus 304. The communication interface 301 is used for data communication between the computer apparatus 300 and an external apparatus. The memory 302 may be used for storing software programs and modules, and the processor 303 may operate the software programs and modules stored in the memory 302, for example, the software programs of the corresponding operations in the foregoing method embodiments.
Alternatively, the processor 303 may call the software programs and modules stored in the memory 302 to perform the following operations: acquiring information content; determining an information subject label of the information content; judging whether a target theme label matched with the information theme label exists in a first image library or not, wherein the image in the first image library is marked with the corresponding theme label; if the first image library has a target subject label matched with the information subject label, taking an image corresponding to the target subject label as a cover image of the information content.
It will be understood by those skilled in the art that all or part of the steps of the methods of the above embodiments may be performed by instructions or by associated hardware controlled by the instructions, which may be stored in a computer readable storage medium and loaded and executed by a processor.
To this end, embodiments of the present application provide a computer-readable storage medium, in which a plurality of computer programs are stored, and the computer programs can be loaded by a processor to perform the steps in any of the artificial intelligence based information mapping methods provided by the embodiments of the present application. The above operations can be implemented in the foregoing embodiments, and are not described in detail herein.
Wherein the storage medium may include: read Only Memory (ROM), random Access Memory (RAM), magnetic or optical disk, and the like.
Since the computer program stored in the storage medium can execute the steps in any artificial intelligence based information mapping method provided in the embodiments of the present application, the beneficial effects that any artificial intelligence based information mapping method provided in the embodiments of the present application can achieve can be achieved, for details, see the foregoing embodiments, and are not described herein again.
Embodiments of the present application also provide a computer program product including computer instructions stored in a computer-readable storage medium. The processor of the computer device reads the computer instruction from the computer-readable storage medium, and executes the computer instruction, so that the computer device executes a corresponding flow in any artificial intelligence-based information mapping method in the embodiment of the present application, which is not described herein again for brevity.
Embodiments of the present application further provide a computer program, where the computer program includes computer instructions, and the computer instructions are stored in a computer-readable storage medium. The processor of the computer device reads the computer instruction from the computer-readable storage medium, and executes the computer instruction, so that the computer device executes a corresponding flow in any artificial intelligence-based information mapping method in the embodiment of the present application, which is not described herein again for brevity.
The information mapping method, the client, the server, the rights incentive system and the storage medium based on artificial intelligence provided by the embodiment of the present application are introduced in detail, and specific examples are applied in the present application to explain the principle and the implementation of the present application, and the description of the above embodiments is only used to help understanding the method and the core idea of the present application; meanwhile, for those skilled in the art, according to the idea of the present application, the specific implementation manner and the application scope may be changed, and in summary, the content of the present specification should not be construed as a limitation to the present application.
Claims (19)
1. An information mapping method based on artificial intelligence, which is characterized in that the method comprises the following steps:
acquiring information content;
determining an information subject label of the information content based on a preset algorithm model;
judging whether a target theme label matched with the information theme label exists in a first image library or not, wherein the image in the first image library is marked with a corresponding theme label;
if the first image library has a target subject label matched with the information subject label, taking an image corresponding to the target subject label as a cover image of the information content.
2. The artificial intelligence based information mapping method of claim 1, wherein said determining the information subject label of the information content based on the predetermined algorithm model comprises:
and acquiring an information title text corresponding to the information text data in the information content, and processing the information title text based on the preset algorithm model to determine the information subject label of the information content.
3. The artificial intelligence based information mapping method as claimed in claim 2, wherein the predetermined algorithm model includes an entity extraction model and a keyword extraction model, and the processing of the information title text based on the predetermined algorithm model to determine the information subject label of the information content includes:
processing the information title text based on an entity extraction model to obtain an entity type label of the information content, wherein the entity extraction model is used for extracting company, industry and name information in the information title text;
processing the information title text based on a keyword extraction model to obtain a keyword label of the information content, wherein the keyword extraction model is used for extracting industry keyword information in the information title text;
and determining the information subject label of the information content according to the entity type label and the keyword label.
4. The artificial intelligence based information mapping method of claim 3, wherein said determining the information subject label of the information content according to the entity type label and the keyword label comprises:
generating a tag list according to the entity type tag and the keyword tag;
and determining the first ranked tag in the tag list as the information subject tag of the information content.
5. The artificial intelligence-based information mapping method of claim 1, wherein said obtaining information content comprises:
crawling information content based on an information content source, and warehousing the information content into an information content library; or
Responding to a warehousing request aiming at the information content sent by the background management equipment, acquiring the information content, and warehousing the information content into an information content library.
6. The artificial intelligence-based information mapping method of claim 5, further comprising, after said obtaining information content:
performing information auditing on the information content based on a preset auditing rule, wherein the preset auditing rule at least comprises sensitive word matching and filtering rule verification;
and if the information content does not hit the sensitive words and the information content does not hit the filtering rules, executing a step of determining the information subject label of the information content based on a preset algorithm model.
7. The artificial intelligence based information mapping method of claim 6, wherein said method further comprises:
and if the information content hits the sensitive words and/or the information content hits the filtering rules, generating first prompt information which is not approved, and sending the information content and the first prompt information to the background management equipment.
8. The artificial intelligence based information mapping method of claim 1, wherein said method further comprises:
if the target theme tag matched with the information theme tag does not exist in the first image library, acquiring a random image from a second image library as a cover image of the information content, wherein the image in the first image library is not labeled with the theme tag.
9. The artificial intelligence based information mapping method of claim 1, wherein said method further comprises:
when a cover page map configuration request aiming at the information content sent by a background management device is received, determining the cover page map of the information content according to configuration information carried in the cover page map configuration request.
10. The artificial intelligence-based information mapping method of claim 1, wherein said method further comprises:
when a cover page image display request aiming at the information content sent by a client is received, sending the cover page image of the information content to the client so as to display the cover page image of the information content on an information preview interface of the client.
11. The artificial intelligence based information mapping method of claim 10, wherein said method further comprises:
when receiving a trigger operation aiming at the cover page picture sent by the client, sending an information stream of the information content to the client so as to display the information stream of the information content on an information preview interface of the client, wherein the information stream comprises at least one of information text data and information multimedia data corresponding to the information content.
12. The artificial intelligence-based information mapping method of claim 1, wherein said method further comprises:
acquiring matching record data in a preset time period, wherein each matching record comprises paired information subject labels and a determination mode of a cover map corresponding to the information subject labels;
counting the occurrence frequency of each information subject label in the preset time period according to the matching image recording data to obtain label statistical data;
according to the image matching record data, counting the data proportion of the determined mode of each cover image in the preset time period to obtain image matching statistical data;
and sending the tag statistical data and the map matching statistical data to a background management device so as to display the tag statistical data and the map matching statistical data on the background management device.
13. The artificial intelligence-based information mapping method of claim 12, wherein said method further comprises:
when an update request aiming at the first gallery sent by the background management equipment is received, at least one of the pictures and the theme labels in the first gallery is updated according to the update request.
14. The artificial intelligence based information charting method of claim 13, wherein said method further comprises:
and if detecting that the same theme label as the label to be updated carried in the updating request exists in the first gallery, generating second prompt information representing repeated labels, and refusing to update the first gallery.
15. The method as claimed in claim 1, wherein if there is a target topic tag matching the information topic tag in the first gallery, using a picture corresponding to the target topic tag as a cover picture of the information content comprises:
if a target theme label matched with the information theme label exists in the first image library and a plurality of images corresponding to the target theme label exist, selecting one image from the plurality of images corresponding to the target theme label as the cover image of the information content.
16. An information mapping device based on artificial intelligence, the device comprising:
an acquisition unit for acquiring information content;
the determining unit is used for determining the information subject label of the information content based on a preset algorithm model;
the judging unit is used for judging whether a target theme label matched with the information theme label exists in a first image library or not, wherein the image in the first image library is marked with the corresponding theme label;
and the processing unit is used for taking the picture corresponding to the target theme label as a cover picture of the information content if the target theme label matched with the information theme label exists in the first image library.
17. A computer device, comprising a processor and a memory, wherein the memory stores a computer program, and the processor is configured to execute the artificial intelligence based information mapping method according to any one of claims 1-15 by calling the computer program stored in the memory.
18. A computer-readable storage medium, wherein the computer-readable storage medium stores a computer program adapted to be loaded by a processor to perform the artificial intelligence based information mapping method according to any one of claims 1-15.
19. A computer program product comprising computer instructions that, when executed by a processor, perform the artificial intelligence based information mapping method according to any of claims 1-15.
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| PCT/CN2022/103226 WO2024000554A1 (en) | 2022-07-01 | 2022-07-01 | Artificial intelligence-based method for configuring image for information, device, medium, and program product |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115298660A true CN115298660A (en) | 2022-11-04 |
Family
ID=83819277
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202280002306.3A Pending CN115298660A (en) | 2022-07-01 | 2022-07-01 | Information mapping method, apparatus, medium, and program product based on artificial intelligence |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN115298660A (en) |
| WO (1) | WO2024000554A1 (en) |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104239535B (en) * | 2014-09-22 | 2018-04-27 | 重庆邮电大学 | A kind of method, server, terminal and system for word figure |
| CN108733779B (en) * | 2018-05-04 | 2022-10-04 | 百度在线网络技术（北京）有限公司 | Text matching method and device |
| CN112749291A (en) * | 2020-08-25 | 2021-05-04 | 腾讯科技（深圳）有限公司 | Image-text matching method and device |
-
2022
- 2022-07-01 CN CN202280002306.3A patent/CN115298660A/en active Pending
- 2022-07-01 WO PCT/CN2022/103226 patent/WO2024000554A1/en unknown
Also Published As
| Publication number | Publication date |
|---|---|
| WO2024000554A1 (en) | 2024-01-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP5256293B2 (en) | System and method for including interactive elements on a search results page | |
| CN108694223B (en) | User portrait database construction method and device | |
| US20210097089A1 (en) | Knowledge graph building method, electronic apparatus and non-transitory computer readable storage medium | |
| CN110309264B (en) | Method and device for acquiring geographic product data based on knowledge graph | |
| US8707167B2 (en) | High precision data extraction | |
| CN109033358B (en) | Method for associating news aggregation with intelligent entity | |
| US20090276716A1 (en) | Content Adaptation | |
| CN101568921A (en) | Dynamic pricing models for digital content | |
| EP3563240B1 (en) | Systems and methods for harvesting data associated with fraudulent content in a networked environment | |
| CN106354861A (en) | Automatic film label indexing method and automatic indexing system | |
| JP2013541793A (en) | Multi-mode search query input method | |
| CN113079417B (en) | Method, device and equipment for generating bullet screen and storage medium | |
| US20170132267A1 (en) | Pushing system and method based on natural information recognition, and a client end | |
| CN103838862B (en) | Video searching method, device and terminal | |
| CN107977678A (en) | Method and apparatus for output information | |
| CN109308338B (en) | Method and equipment for searching and displaying information in application program | |
| CN105893584A (en) | Method, client and system for displaying website label of favorites | |
| CN111831885B (en) | Internet information retrieval system and method | |
| CN115298660A (en) | Information mapping method, apparatus, medium, and program product based on artificial intelligence | |
| US20140013194A1 (en) | System and method for collecting financial information over a global communications network | |
| CN112487277B (en) | Data distribution method and device, readable storage medium and electronic equipment | |
| CN113407678A (en) | Knowledge graph construction method, device and equipment | |
| CN115630170B (en) | Document recommendation method, system, terminal and storage medium | |
| KR101696026B1 (en) | Design trend information supplying method by open API | |
| CN115984004A (en) | Information association method, device, equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |