Information processing method and device, storage medium and electronic equipment

Abstract

The invention discloses an information processing method and device, a storage medium and electronic equipment. The method comprises the steps of obtaining original description information of a supply object, determining supply expression data based on the original description information of the supply object, and determining a first data feature of the supply expression data; obtaining original demand information, determining demand expression data based on the original demand information, and determining a second data feature of the demand expression data; and performing supply and demand prediction on the first data feature of the supply expression data and the second data feature of the demand expression data based on a supply and demand prediction model to obtain a supply and demand prediction result. The supply expression data and the demand expression data in the same semantic space are obtained, high-precision matching is carried out, supply and demand prediction is carried out through the first data feature of the supply expression data and the second data feature of the demand expression data, and the accuracy of a supply and demand prediction result is improved.

Description

Technical Field

[0001] The present invention relates to the field of information processing technology, and in particular to an information processing method, device, storage medium and electronic device. Background Art

[0002] In the e-commerce scenario, supply, demand and the relationship between the two play a vital role in the distribution and conversion of online trading items, as well as offline procurement and sales. When supply cannot meet demand, users initiate searches but cannot retrieve the desired products, which not only loses the opportunity to trade, but also seriously affects the user experience and retention. When there is an oversupply but demand cannot accurately describe the items, the trading items cannot be exposed, and users lack awareness of the platform's product reserves, which is not conducive to promoting the liquidity of trading items.

[0003] In the process of implementing the present invention, it is found that there are at least the following technical problems in the prior art: the supply status and demand satisfaction of trading items are difficult to quantify and predict. Summary of the invention

[0004] The present invention provides an information processing method, device, storage medium and electronic device to semantically unify supply information and demand information for prediction processing, thereby improving the accuracy of prediction results.

[0005] According to one aspect of the present invention, there is provided an information processing method, comprising:

[0006] Acquire original description information of a supply object, determine supply expression data based on the original description information of the supply object, and determine a first data feature of the supply expression data;

[0007] Acquire original demand information, and determine demand expression data based on the original demand information, and determine a second data feature of the demand expression data;

[0008] Based on the supply and demand forecasting model, supply and demand forecasting is performed on the first data feature of the supply expression data and the second data feature of the demand expression data to obtain a supply and demand forecasting result.

[0009] Optionally, determining the supply expression data based on the original description information of the supply object includes: extracting product entities and attribute entities in the original description information, and forming structured data based on the product entities and attribute entities; and forming supply expression data based on the structured data.

[0010] Optionally, extracting the product entity and the attribute entity in the original description information includes:

[0011] The entity words in the original description information are matched in a preset product entity knowledge graph to obtain a product entity; the entity words in the original description information are matched in a preset attribute entity knowledge graph to obtain an attribute entity.

[0012] Optionally, after extracting the product entity and attribute entity from the original description information, it also includes: performing semantic processing on the product entity and / or the attribute entity based on a lexical relationship graph to obtain processed product entities and / or attribute entities, wherein the lexical relationship graph includes one or more of the following: a synonym relationship graph, a hyponym relationship graph, and a product attribute relationship graph.

[0013] Optionally, the forming of supply expression data based on structured data of multiple supply objects includes: for any of the supply objects, combining a product entity in the structured data with at least one attribute entity to obtain at least one supply expression data; and deduplicating the supply expression data.

[0014] Optionally, after obtaining the original demand information, it also includes:

[0015] The original demand data is subjected to error correction processing and text verification, and in the case of file verification failure, the original demand data is discarded, wherein the file verification includes one or more of the following: integrity verification and banned word recognition.

[0016] Optionally, supply and demand forecasting is performed on the first data feature of the supply expression data and the second data feature of the demand expression data based on a supply and demand forecasting model to obtain a supply and demand forecasting result, including:

[0017] The supply expression data and the demand expression data are matched to determine the matching supply expression data and demand expression data; the first data feature and the second data feature respectively corresponding to the matching supply expression data and the demand expression data are input into the supply and demand forecasting model to obtain the supply and demand forecasting results of the supply expression data and the demand expression data.

[0018] Optionally, the first data features include supply statistical features and supply evaluation features, and the second data features include demand statistical features and demand evaluation features.

[0019] Optionally, the supply and demand forecast result includes a supply type and a demand type, the supply type includes at least two supply grade types, and the demand type includes at least two grade types.

[0020] Optionally, the original description information of the supply object is the original description information within a preset time period, and the original demand information is the original demand information within the preset time period, and the preset time period is determined based on a forecast time range for supply and demand forecasting.

[0021] According to another aspect of the present invention, there is provided an information processing device, comprising:

[0022] A supply information processing module, which obtains original description information of a supply object, determines supply expression data based on the original description information of the supply object, and determines a first data feature of the supply expression data;

[0023] A demand information processing module, used to obtain original demand information, determine demand expression data based on the original demand information, and determine a second data feature of the demand expression data;

[0024] The supply and demand forecasting module is used to perform supply and demand forecasting on the first data feature of the supply expression data and the second data feature of the demand expression data based on the supply and demand forecasting model to obtain a supply and demand forecasting result.

[0025] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0026] at least one processor; and

[0027] a memory communicatively connected to the at least one processor; wherein,

[0028] The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can perform the information processing method described in any embodiment of the present invention.

[0029] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the information processing method described in any embodiment of the present invention when executed.

[0030] The technical solution of the embodiment of the present invention semantically processes the original description information and the original demand information of the supply object to obtain supply expression data and demand expression data in the same semantic space. High-precision matching of the supply expression data and the demand expression data can be achieved in the semantic space, and supply and demand forecasts are performed through the first data features of the supply expression data and the second data features of the demand expression data, thereby improving the accuracy of the supply and demand forecast results.

[0031] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0033] Figure 1 is a flow chart of an information processing method provided by an embodiment of the present invention;

[0034] Figure 2 is a schematic diagram of the processing process of the original description information provided by the embodiment of the invention;

[0035] Figure 3 It is a schematic diagram of the processing process of the original demand information provided by an embodiment of the present invention;

[0036] Figure 4 is a schematic diagram of the relationship between supply information and demand information provided by an embodiment of the present invention;

[0037] Figure 5 is a flow chart of an information processing method provided by an embodiment of the present invention;

[0038] Figure 6 is a structural schematic diagram of an information processing device provided by an embodiment of the present invention;

[0039] Figure 7 It is a structural schematic diagram of an electronic device implementing an embodiment of the present invention. DETAILED DESCRIPTION

[0040] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0041] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0042] Figure 1 is a flowchart of an information processing method provided by an embodiment of the present invention. This embodiment is applicable to the case of supply and demand forecasting. The method can be executed by an information processing device. The information processing device can be implemented in the form of hardware and / or software. The information processing device can be configured in electronic devices such as mobile phones, computers, servers, etc. Figure 1 As shown, the method includes:

[0043] S110. Acquire original description information of a supply object, determine supply expression data based on the original description information of the supply object, and determine a first data feature of the supply expression data.

[0044] S120, obtaining original demand information, and determining demand expression data based on the original demand information, and determining a second data feature of the demand expression data;

[0045] S130. Perform supply and demand forecasting on the first data feature of the supply expression data and the second data feature of the demand expression data based on a supply and demand forecasting model to obtain a supply and demand forecasting result.

[0046] In this embodiment, the supply object may be a transaction object that can be provided by the supplier. The type of supply object is not limited here, including but not limited to physical items and virtual items. Among them, the supplier may be a merchant or a trading platform. The original description information of each supply object that can be provided by the supplier is obtained. The original description information of the supply object may be sku information, including but not limited to text information such as the title and item description of the supply object. The original description information of the supply object may be pre-stored in the database, and the corresponding original description information in the database may be updated according to the changes of the supply object. Among them, the update of the database includes but is not limited to the addition, deletion and partial update of the original description information. The supply information is reflected through the original description information of the supply object, providing a data basis for supply and demand forecasting.

[0047] The user's needs are reflected by the access and query of the supply object. Accordingly, the user's query and access to any transaction object are obtained to obtain the original demand information. In some embodiments, the query information entered by the user through the search box is received and used as the original demand information; in some embodiments, the query tag is displayed through the interactive page, and when the user queries the supply object by clicking the query tag, the received query tag is used as the original demand information; in some embodiments, the relevant information of the supply object (such as images, audio and video or description information, etc.) is displayed through the interactive interface, and the supply object display page is accessed in response to the trigger operation of the relevant information of the supply object, and the original description information (such as sku information) of the supply object can be used as the original demand information.

[0048] The original description information is stored in a database, which may include the original description information and the timestamp corresponding to the original description information. The original demand information reflects the user's demand for the supply object, providing a data basis for supply and demand forecasting.

[0049] In response to the triggering operation of the supply and demand forecast, or when the triggering conditions (such as periodic conditions) of the supply and demand forecast are met, the original description information and the original demand information of the supply object are read from the database to perform supply and demand forecasting. The database includes the supply time information of the supply object, and accordingly, the original description information of the supply object within the preset time period can be obtained. Optionally, the original description information of the supply object is the original description information within the preset time period, and the original demand information is the original demand information within the preset time period. The preset time period is determined based on the forecast time range for performing the supply and demand forecast, and the length of the preset time period is positively correlated with the length of the forecast time range for performing the supply and demand forecast.

[0050] In some embodiments, different preset time periods are set to obtain supply and demand forecast results of different forecast durations, where different forecast durations include but are not limited to the next day, the next week, and the next month. By obtaining supply and demand forecast results of different forecast durations, the supply quantity of the supply object can be flexibly adjusted to meet user needs and avoid the problem of oversupply. Optionally, multiple supply and demand forecast tasks are triggered in parallel according to different preset time periods, and multiple supply and demand forecast tasks are executed in parallel to speed up the efficiency of supply and demand forecasts of multiple forecast durations.

[0051] The original description information and the original demand information are semantically processed and unified into the same semantic space, so as to establish an association relationship between supply and demand. Specifically, the original description information is semantically processed to obtain supply expression data, and the original demand information is semantically processed to obtain demand expression data. The supply expression data and the demand expression data are located in the same semantic space and have the same data structure, which can achieve the matching of the supply expression data and the demand expression data, provide the matching accuracy of supply and demand, and avoid the difficulty of matching caused by different expression methods and expression dimensions of supply and demand.

[0052] Among them, the supply expression data and the demand expression data include product entities and attribute entities respectively, and there is at least one attribute entity. The product entities in the supply expression data and the demand expression data are matched respectively, and the attribute entities are matched. When the product entity and the attribute entity are the same, the supply expression data and the demand expression data are characterized as the same target expression data. Supply and demand prediction can be performed on the matched supply expression data and demand expression data to determine the supply and demand prediction results of the target expression data.

[0053] For the original description information, supply expression data is determined based on the original description information of the supply object, including: extracting product entities and attribute entities in the original description information, and forming structured data based on the product entities and attribute entities; forming supply expression data based on the structured data.

[0054] The original description information of the supply object includes multiple information, and entity word recognition is performed on the original description information to obtain product entities and attribute entities in the original description information. Optionally, the original description information is segmented, and the entity words in the segmented words are identified, and the identified entity words are classified to obtain product entities and attribute entities. Optionally, product entity word recognition is performed on the original description information to obtain product entities, and attribute entity word recognition is performed on the original description information to obtain attribute entities. Among them, before product entity word recognition and attribute entity word recognition, it also includes segmentation of the original description information, deduplication of the segmented words, and elimination of segmented words other than entity words. Stop words, etc. can be deleted to avoid computing power and time consumption caused by processing non-entity words. Specifically, a product entity word set and an attribute entity word set can be set in advance, and product entities and attribute entities can be obtained by matching entity words in the product entity word set and the attribute entity word set.

[0055] Optionally, extracting the product entity and attribute entity from the original description information includes: matching the entity words in the original description information in a preset product entity knowledge graph to obtain the product entity; matching the entity words in the original description information in a preset attribute entity knowledge graph to obtain the attribute entity. The product entity knowledge graph and the attribute entity knowledge graph are preset and can be generated based on the product entity and attribute entity of the historical supply object. The product entity knowledge graph and the attribute entity knowledge graph can be updated as needed. Exemplarily, in the case of a newly added supply object, it is determined whether the product entity in the newly added supply object is included in the product entity knowledge graph, and whether the attribute entity in the newly added supply object is included in the attribute entity knowledge graph. If not, the product entity knowledge graph is updated based on the product entity of the newly added supply object, and / or the attribute entity knowledge graph is updated based on the attribute entity of the newly added supply object to improve the accuracy of the knowledge graph.

[0056] If any entity word is successfully matched in the product entity knowledge graph, the entity word is determined to be a product entity, and if any entity word is successfully matched in the attribute entity knowledge graph, the entity word is determined to be an attribute entity. For example, the original description information can be "neck hanging fan head-mounted neck hanging bladeless small fan", and accordingly, the identified product entity is "fan", and the attribute entities are "neck hanging", "head-mounted", and "bladeless".

[0057] The identified product entities and attribute entities are stored as structured data. The structured template is called, and the identified product entities and attribute entities are added to the structured template to form structured data. Among them, the structured template can be: {product center word: X, attribute word: X}, wherein the "product center word" and "attribute word" in the above template can be identified by other characters or strings. For example, the structured template can also be: {p: X, t: X}. Exemplarily, the structured data formed by the above original description information can be {product center word: fan, attribute words: neck-hanging, head-mounted, leafless}.

[0058] In order to avoid the inclusion of semantically repeated entity words or inaccurate entity words in the identified attribute entities, the identified attribute entities can be semantically normalized, the attribute entities can be simplified, and incorrect attribute entities can be eliminated. Accordingly, after extracting the product entity and attribute entity in the original description information, it also includes: semantically processing the product entity and / or the attribute entity based on a vocabulary relationship map to obtain a processed product entity and / or attribute entity, wherein the vocabulary relationship map includes one or more of the following: a synonym relationship map, a hyponym relationship map, and a product attribute relationship map.

[0059] Among them, the synonym relationship map includes the synonym relationship between entity words, and the synonym relationship map connects two entity words that are synonyms with each other through a line segment to represent that the two entity words have a synonym relationship. For any supply object, the multiple attribute entities identified in the original description information of the supply object are matched in the synonym relationship map respectively, and it is determined whether any two attribute entities are connected by a line segment in the synonym relationship map. If so, the two attribute entities are determined to be synonyms, and one of the two attribute entities that are synonyms is retained. For two attribute entities that are synonyms, the attribute entity with a high priority can be retained, or the attribute entity with a high frequency of occurrence can be included. Similarly, when there are two or more identified product entities, the synonym recognition of the product entity can be performed through the synonym relationship map. The recognition process is the same as the above process and will not be repeated here. By performing synonym recognition on product entities and / or attribute entities, entity words with repeated semantics can be removed and structured data can be simplified.

[0060] The hyponymy graph includes the hyponymy of entity words, and the hyponymy graph represents two entity words with a hyponymy relationship through a specific identifier. Exemplarily, the specific identifier can be an arrow, for example, A→B, which can indicate that A is the hypernym of B; Exemplarily, the specific identifier can be >, for example, A>B, which can indicate that A is the hypernym of B. For any supply object, multiple attribute entities identified in the original description information of the supply object are matched in the hyponymy graph respectively to determine whether any two attribute entities have a hyponymy relationship in the hyponymy graph, and specifically determine whether any two attribute entities are connected by a specific identifier. If so, it is determined that they have a hyponymy relationship. Optionally, in the case where two attribute entities have a hyponymy relationship, the processing method of the hyponymy entity is preset, for example, the subordinate attribute entity is retained and the superior attribute entity is eliminated; or, the superior attribute entity is retained and the subordinate attribute entity is eliminated, which can be determined here according to the preset processing method. For example, for the attribute entity of the fan, "neck hanging" is a hyponym of "head-mounted". According to the preset processing method, "neck hanging" can be retained and "head-mounted" can be eliminated. Similarly, when there are two or more identified product entities, the hyponymy relationship of the product entities can be identified through the hyponymy relationship map. The identification process is the same as the above process and will not be repeated here. By identifying the hyponymy of product entities and / or attribute entities, unnecessary entity words can be eliminated from the entity words of the hyponymy relationship to simplify the structured data.

[0061] The product attribute relationship map includes the correspondence between product entities and attribute entities. One product entity can correspond to multiple attribute entities, and entity words with corresponding relationships can be connected by line segments. For any supply object, the attribute entity and product entity identified in the original description information of the supply object are matched in the product attribute relationship map respectively to determine whether the attribute entity and the product entity have a corresponding relationship in the product attribute relationship map, that is, to determine whether the attribute entity and the product entity are connected by a line segment. If so, it is determined that the two have a corresponding relationship. If not, it is determined that the two do not have a corresponding relationship. The fact that the product entity and the attribute entity do not have a corresponding relationship indicates that the attribute entity cannot accurately express the product entity, and the attribute entity that does not have a corresponding relationship with the product entity is eliminated. By identifying the corresponding relationship between the product entity and the attribute entity, inaccurate attribute entities are eliminated, and the accuracy of the attribute entity's expression of the product entity is improved.

[0062] In some embodiments, since the original description information of the supply object includes a large amount of information, the identified attribute entities are relatively large. Before generating structured data, it also includes identifying the core attribute entity among the multiple attribute entities, and accordingly, generating structured data based on the product entity and the core attribute entity. By eliminating non-core attribute entities, the amount of information to be processed is reduced, and the interference of non-core attributes is reduced. Optionally, the identified attribute entities are sorted according to the priority of the attribute entity set in advance, and the attribute entities located in the preset sorting range are determined as core attribute entities according to the sorting, for example, the attribute entities located in the first n positions of the sorting can be determined as core attribute entities, or the attribute entities located in the first m% can be determined as core attribute entities, where m and n are positive integers, respectively, and the specific values ​​are determined according to demand. It can be understood that different attribute entity priorities are set for supply objects of different categories, and the attribute entity priority is called according to the category of the supply object. Optionally, the attribute entities are sorted according to the historical demand frequency of the attribute entity to determine the core attribute entity according to the sorting. Among them, according to the attribute entity entered by the user in the historical search and access process, the historical demand frequency of the attribute entity is counted, and further, the historical demand frequency of each attribute entity under each category is determined. Get the historical demand frequency of the attribute entity at the current moment and sort the attribute entity. Optionally, call the pre-set attribute entity screening model, input the identified product entity and attribute entity into the attribute entity screening model, and obtain the core attribute entity output by the attribute entity screening model. The attribute entity screening model has the function of screening the core attribute entity of any product entity, and is pre-trained based on sample attribute entities and core attribute entity labels.

[0063] Through the above processing method, the product entity and attribute entity in the original description information are extracted, and the product entity and attribute entity are semantically normalized and the core entity is identified, and structured data is formed based on the processed product entity and core attribute entity. The structured data of the original description information of each supply object is determined respectively, wherein the original description information of different supply objects can form the same structured data, and illustratively, the same structured data can be formed for the same type of items from different manufacturers or origins.

[0064] For each structured data, the structured data includes a product entity and at least one attribute entity, one or more attribute entities are combined to obtain an attribute entity combination, which is used to express the product entity to form supply expression data. Specifically, the supply expression data is formed based on the structured data of multiple supply objects, including: for any of the supply objects, combining the product entity in the structured data with at least one attribute entity to obtain at least one supply expression data; and performing deduplication processing on the supply expression data.

[0065] For example, for the structured data {product core word: fan, attribute word: neck hanging, portable}, which includes two attribute entities, one or more of the two attribute entities are combined to obtain the following supply expression data: {product core word: fan, attribute word: neck hanging}, {product core word: fan, attribute word: portable}, {product core word: fan, attribute word: neck hanging, portable}. By forming supply expression data of different ranges, supply objects can be clustered in different ranges, which facilitates statistics of supply objects.

[0066] For example, see Figure 2 , Figure 2 Schematic diagram of the processing of original description information provided by the embodiment of the invention. SKU1, SKU2, SKU3 and SKU4 are original description information of different supply objects, AU1 and AU2 are structured data, and KU1, KU2 and KU3 are supply expression data. Figure 2 It can be seen that the original description information of different supply objects can obtain the same structured data, the same structured data can obtain different supply expression data, and different structured data can obtain the same supply expression data.

[0067] The original demand information is subjected to entity word recognition and semantic normalization processing to obtain demand expression data, wherein the processing process of entity word recognition and semantic normalization processing is the same as the processing process of the original description data, which will not be repeated here. It can be understood that the product entity and attribute entity are recognized for the original demand information, and semantic normalization processing is performed on the product entity and the attribute entity respectively to obtain structured data. In order to ensure the accuracy of the demand expression data, the structured data is directly used as the demand expression data, and the structured data is no longer combined.

[0068] On the basis of the above embodiment, since the original demand information is input by the user, in order to ensure the validity and accuracy of the original demand information, the original demand information is preprocessed after obtaining the original demand information and before performing entity word recognition and semantic normalization processing. Specifically, the preprocessing can be error correction processing and text verification of the original demand data, and in the case of file verification failure, the original demand data is discarded, wherein the file verification includes one or more of the following: integrity verification and banned word recognition.

[0069] Among them, error correction processing is used to correct typos in the original demand information. For example, "Lianhua Qingwen" can be corrected to "Lianhua Qingwen". Specifically, an error correction model can be obtained by training the wrong words and the corresponding correct words, and the original demand information can be input into the error correction model for processing to obtain the original demand information after error correction processing. Error correction processing improves the accuracy of the original demand information and avoids the problem of inaccurate demand expression caused by wrong words.

[0070] The text verification process may be located after the error correction process, and the original demand information after the error correction process is subjected to text verification. Specifically, integrity verification and banned word recognition are performed. The integrity verification and banned word recognition may be performed in parallel or in any order, which is not limited here.

[0071] Among them, the banned word identification can be to match each word in the original demand information in the banned word set. If any word is successfully matched in the banned word set, it is determined that the original demand information includes the banned word, and the original demand information is removed.

[0072] The integrity check can be implemented through an integrity check model to determine whether the original demand information is complete. If it is incomplete, the original demand information is discarded. If it is complete, the original demand information is retained. The integrity check model can be a classification model, and the output result is complete or incomplete. By performing integrity check on the original demand information, the integrity of the original demand information is guaranteed to avoid the situation where the incomplete original demand information cannot accurately express the user's needs. For example, see Figure 3 , Figure 3It is a schematic diagram of the processing process of original demand information provided by an embodiment of the present invention.

[0073] After semantic processing of the original demand information and the original description information, the supply expression data and the demand expression data in the same semantic space are obtained, and the first data feature of the supply expression data and the second data feature of the demand expression data are obtained to perform supply and demand forecasting. Among them, the first data feature includes supply statistical features and supply evaluation features, and the second data feature includes demand statistical features and demand evaluation features. Among them, the supply statistical features include the first statistical index, and the demand statistical features include the second statistical index. The first statistical index includes but is not limited to the number of supply objects that meet the supply expression data, the number of brands of supply objects, the type and number of stores of supply objects, etc. The second statistical index includes but is not limited to the number of exposures, the number of clicks, the number of orders, etc. of products that meet the demand expression data. It can be understood that in the database of the original description information of the supply object, the number, brand, type, etc. of the supply object is also included, and the first statistical index can be used to query in the database to obtain the index data of the first statistical index. Similarly, when the supply object is displayed to the client and the client accesses and queries are received, the exposure, clicks, and ordering data of the supply object are recorded. Correspondingly, the second statistical index is queried in the above-mentioned recorded information to obtain the index data of the second statistical index. By presetting the second statistical indicator and the first statistical indicator, indicator data corresponding to the second statistical indicator and the first statistical indicator are determined respectively, thereby simplifying the process of determining data features.

[0074] The supply evaluation feature can be obtained by creating a first portrait feature for the supply object that meets the supply expression data and evaluating the first portrait feature. The demand evaluation feature can be obtained by creating a second portrait feature for the supply object that meets the demand expression data in the displayed supply object and evaluating the second portrait feature. Among them, the features in the first portrait feature and the second portrait feature include but are not limited to the brand, material, function, etc. of the supply object. The portrait features of supply objects of different categories are different and can be set according to demand. Optionally, the feature type included in the portrait feature in each category is pre-set, and the corresponding feature type is called according to the product entity included in the supply expression data and the demand expression data. For the supply expression data, the first portrait feature is obtained according to the determined feature type and the feature value of the above feature type determined based on the supply object of the demand expression data; similarly, for the demand expression data, the second portrait feature is obtained according to the determined feature type and the feature value of the above feature type determined based on the displayed supply object that meets the demand expression data.

[0075] The first image feature is evaluated and processed to obtain a supply evaluation feature, and the second image feature is evaluated and processed to obtain a demand evaluation feature. The supply evaluation feature and the demand evaluation feature can be data of 0-1, respectively, used to express the supply quality and the demand quality. In some embodiments, the first image feature or the second image feature can be evaluated and processed by a pre-set evaluation model. The evaluation model can be a neural network model trained based on the image features and the evaluation labels.

[0076] The supply expression data and the demand expression data are located in the same semantic space, and the supply expression data and the demand expression data can be matched. The matched supply expression data and the demand expression data belong to the same target expression data. The supply and demand prediction of the target expression data can be performed based on the first data feature of the supply expression data and the second data feature of the demand expression data to obtain the supply and demand prediction result of the target expression data. Among them, the matching of the supply expression data and the demand expression data includes the matching of product entities and the matching of attribute entities. The product entities and attribute entities included in the matched supply expression data and demand expression data are the same.

[0077] Optionally, supply and demand prediction is performed on the first data feature of the supply expression data and the second data feature of the demand expression data based on a supply and demand prediction model to obtain a supply and demand prediction result, including: matching the supply expression data and the demand expression data to determine matching supply expression data and demand expression data; inputting the first data feature and the second data feature respectively corresponding to the matching supply expression data and demand expression data into the supply and demand prediction model to obtain supply and demand prediction results for the supply expression data and the demand expression data.

[0078] The supply and demand forecasting model may be a neural network model, which is an end-to-end forecasting model that can be trained using the first data feature and the second data feature of the sample expression data and the supply and demand labels. The structure and training process of the supply and demand forecasting model are not limited here.

[0079] The supply and demand forecast results include supply types and demand types, the supply types include at least two supply level types, and the demand types include at least two level types. In some embodiments, the supply types include high supply and low supply, and the demand types include high demand and low demand. Accordingly, the supply and demand forecast results include high demand and high supply, high demand and low supply, low demand and high supply, and low demand and low supply. For example, see Figure 4 , Figure 4 It is a schematic diagram of the relationship between supply information and demand information provided by an embodiment of the present invention.

[0080] The supply and demand forecast results in this embodiment can express the relationship between supply and demand. For example, high demand and low supply indicate that supply is lower than demand, low demand and high supply indicate that supply is higher than demand, and high demand and high supply and low demand and low supply indicate that supply and demand are in a relatively balanced state. At the same time, the supply and demand forecast results can also express the satisfaction degree of the supply itself and the heat of the demand itself. For example, high demand and low demand represent different demand heats, and high supply and low supply represent different supply satisfactions.

[0081] By displaying the supply and demand forecast results, a processing strategy for the supply objects can be generated. For example, traffic can be provided to supply objects with high supply and low demand, and supply can be strengthened for supply objects with high demand and low supply, so as to meet or stimulate user demand.

[0082] The technical solution of this embodiment obtains supply expression data and demand expression data in the same semantic space by semantically processing the original description information and original demand information of the supply object. High-precision matching of the supply expression data and the demand expression data can be achieved in the semantic space, and supply and demand forecasts are performed through the first data features of the supply expression data and the second data features of the demand expression data, thereby improving the accuracy of the supply and demand forecast results.

[0083] Based on the above embodiments, the present invention also provides a preferred example of an information processing method. Figure 5 , Figure 5It is a flowchart of an information processing method provided by an embodiment of the present invention. A cognitive map is pre-set to provide a basis for processing the original description information and the original demand information. Supply analysis is performed on the original description information (i.e., SKU), the sku is unified into the standard term space, and the term components are combined within a certain range as supply expression data. The original description information sku is the smallest unit of supply. The description information is stored in the hive table, which includes structured attributes kv and natural language texts such as titles and product descriptions. NER entity recognition is performed on the original description information sku to obtain product entities and attribute entities, and semantic normalization and core attribute screening are performed on the identified entities to obtain structured data AU. The structured data AU contains a central product word (i.e., product entity) and multiple core attribute kvs (attribute entities), which are used to describe the supply object itself. Through core attribute screening, sku that can be replaced by each other are aggregated. The attribute entities in the structured data AU are combined to obtain the supply expression data KU, which is obtained by combining the central product words and core attributes in the AU. Generally speaking, the number of modifying components is 3 or less; it can be represented by structured data or organized into natural language texts such as product phrases / queries, which are called "supply expressions"; the purpose is to aggregate commodities under the expression length conditions that are consistent with habits and cognition, so as to facilitate better statistics and thus observe signals / discover trends. Furthermore, the category to which the supply expression data KU belongs can also be determined, which is used to describe the relationship between supply categories and categories, corresponding to the existing front-end and back-end categories.

[0084] The original description information SKU, the structured data AU and the supply expression data KU all correspond to the first data features, namely, statistical features and portrait features.

[0085] The original demand information is analyzed and unified into the standard term space as the demand expression. For this expression, the statistical characteristics and portrait characteristics such as demand sufficiency are output with the dual dimensions of time and space as statistical units. The demand comes from user behaviors such as query and tag click, but it may not reflect the user's true intention. The query actively input or clicked by the user is recorded as the original query, and the query components are identified and disassembled to finally convert it into demand expression data in the form of term.

[0086] By matching the supply and demand in the term space, the correlation between supply and demand is obtained, i.e., high demand and high supply, high demand and low supply, low demand and high supply, and low demand and low supply. The demand expression data and the supply expression data are searched and matched according to the types of the central product words and the modifiers; if the user's input does not contain product words, only the modifiers are matched, and the product words are treated as wildcards. After the matching is completed, the correlation between the supply and demand of terms in the future time is predicted through the preset supply and demand forecasting model. By performing a four-quadrant analysis on the supply and demand forecasting results, the quadrants corresponding to the terms are obtained for downstream applications.

[0087] Figure 6 Schematic diagram of the structure of an information processing device provided by an embodiment of the present invention. Figure 6 As shown, the device comprises:

[0088] The supply information processing module 210 acquires original description information of the supply object, determines supply expression data based on the original description information of the supply object, and determines a first data feature of the supply expression data;

[0089] The demand information processing module 220 is used to obtain original demand information, determine demand expression data based on the original demand information, and determine a second data feature of the demand expression data;

[0090] The supply and demand prediction module 230 is used to perform supply and demand prediction on the first data feature of the supply expression data and the second data feature of the demand expression data based on the supply and demand prediction model to obtain a supply and demand prediction result.

[0091] The technical solution of this embodiment obtains supply expression data and demand expression data in the same semantic space by semantically processing the original description information and original demand information of the supply object. High-precision matching of the supply expression data and the demand expression data can be achieved in the semantic space, and supply and demand forecasts are performed through the first data features of the supply expression data and the second data features of the demand expression data, thereby improving the accuracy of the supply and demand forecast results.

[0092] Based on the above embodiment, optionally, the supply information processing module 210 is used to:

[0093] Extract product entities and attribute entities from the original description information, and form structured data based on the product entities and attribute entities; and form supply expression data based on the structured data.

[0094] Optionally, the supply information processing module 210 is further used for:

[0095] Matching the entity words in the original description information in a preset product entity knowledge graph to obtain a product entity;

[0096] The entity words in the original description information are matched in a preset attribute entity knowledge graph to obtain attribute entities.

[0097] Optionally, the supply information processing module 210 is further used for:

[0098] The product entity and / or the attribute entity is semantically processed based on a lexical relationship graph to obtain processed product entities and / or attribute entities, wherein the lexical relationship graph includes one or more of the following: a synonym relationship graph, a hyponym relationship graph, and a product attribute relationship graph.

[0099] Optionally, the supply information processing module 210 is further used for:

[0100] For any of the supply objects, the product entity in the structured data is combined with at least one attribute entity to obtain at least one supply expression data; and the supply expression data is deduplicated.

[0101] Based on the above embodiment, optionally, the demand information processing module 220 is used to:

[0102] The original demand data is subjected to error correction processing and text verification, and in the case of file verification failure, the original demand data is discarded, wherein the file verification includes one or more of the following: integrity verification and banned word recognition.

[0103] Based on the above embodiment, optionally, the supply and demand forecasting module 230 is used to:

[0104] The supply expression data and the demand expression data are matched to determine the matching supply expression data and demand expression data; the first data feature and the second data feature respectively corresponding to the matching supply expression data and the demand expression data are input into the supply and demand forecasting model to obtain the supply and demand forecasting results of the supply expression data and the demand expression data.

[0105] Based on the above embodiment, optionally, the first data feature includes supply statistical feature and supply evaluation feature, and the second data feature includes demand statistical feature and demand evaluation feature.

[0106] Based on the above embodiment, optionally, the supply and demand forecast result includes a supply type and a demand type, the supply type includes at least two supply grade types, and the demand type includes at least two grade types.

[0107] Based on the above embodiment, optionally, the original description information of the supply object is the original description information within a preset time period, and the original demand information is the original demand information within the preset time period, and the preset time period is determined based on a forecast time range for supply and demand forecasting.

[0108] The information processing device provided by the embodiment of the present invention can execute the information processing method provided by any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0109] Figure 7 1 is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. The electronic device 10 is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0110] like Figure 7 As shown, the electronic device 10 includes at least one processor 11, and a memory connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 to the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0111] A number of components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0112] The processor 11 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 performs the various methods and processes described above, such as information processing methods.

[0113] In some embodiments, the information processing method may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the information processing method described above may be performed. Alternatively, in other embodiments, the processor 11 may be configured to perform the information processing method in any other appropriate manner (e.g., by means of firmware).

[0114] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0115] The computer programs for implementing the information processing methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer or other programmable information processing device, so that when the computer programs are executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer programs may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0116] An embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to cause a processor to execute an information processing method, the method comprising:

[0117] Obtain original description information of the supply object, determine supply expression data based on the original description information of the supply object, and determine a first data feature of the supply expression data; obtain original demand information, determine demand expression data based on the original demand information, and determine a second data feature of the demand expression data; perform supply and demand forecasting on the first data feature of the supply expression data and the second data feature of the demand expression data based on a supply and demand forecasting model to obtain a supply and demand forecasting result.

[0118] In the context of the present invention, computer-readable storage media can be tangible media, which can contain or store a computer program for use by an instruction execution system, device or equipment or used in combination with an instruction execution system, device or equipment. Computer-readable storage media can include but are not limited to electronic, magnetic, optical, electromagnetic, infrared or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, computer-readable storage media can be machine-readable signal media. More specific examples of machine-readable storage media can include electrical connections based on one or more lines, portable computer disks, hard disks, random access memories (RAM), read-only memories (ROM), erasable programmable read-only memories (EPROM or flash memory), optical fibers, portable compact disk read-only memories (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0119] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0120] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0121] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0122] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0123] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. An information processing method, characterized in that: include: Acquire original description information of a supply object, determine supply expression data based on the original description information of the supply object, and determine a first data feature of the supply expression data; Acquire original demand information, and determine demand expression data based on the original demand information, and determine a second data feature of the demand expression data; Based on the supply and demand forecasting model, supply and demand forecasting is performed on the first data feature of the supply expression data and the second data feature of the demand expression data to obtain a supply and demand forecasting result.

2. The method according to claim 1, characterized in that: The determining of the supply expression data based on the original description information of the supply object comprises: Extracting product entities and attribute entities from the original description information, and forming structured data based on the product entities and attribute entities; Supply expression data is formed based on the structured data.

3. The method according to claim 2, characterized in that The extracting of product entities and attribute entities from the original description information includes: Matching the entity words in the original description information in a preset product entity knowledge graph to obtain a product entity; The entity words in the original description information are matched in a preset attribute entity knowledge graph to obtain attribute entities.

4. The method according to claim 2, characterized in that: After extracting the product entity and the attribute entity in the original description information, the method further includes: The product entity and / or the attribute entity is semantically processed based on a lexical relationship graph to obtain processed product entities and / or attribute entities, wherein the lexical relationship graph includes one or more of the following: a synonym relationship graph, a hyponym relationship graph, and a product attribute relationship graph.

5. The method according to claim 2, characterized in that: The forming of supply expression data based on structured data of multiple supply objects includes: For any of the supply objects, combining a product entity and at least one attribute entity in the structured data to obtain at least one supply expression data; The supplied expression data is deduplicated.

6. The method according to claim 1, characterized in that After obtaining the original demand information, it also includes: The original demand data is subjected to error correction processing and text verification, and in the case of file verification failure, the original demand data is discarded, wherein the file verification includes one or more of the following: integrity verification and banned word recognition.

7. The method according to claim 1, characterized in that Based on the supply and demand forecasting model, supply and demand forecasting is performed on the first data feature of the supply expression data and the second data feature of the demand expression data to obtain a supply and demand forecasting result, including: Matching the supply expression data with the demand expression data to determine matching supply expression data and demand expression data; The first data feature and the second data feature respectively corresponding to the matching supply expression data and the demand expression data are input into the supply and demand prediction model to obtain supply and demand prediction results of the supply expression data and the demand expression data.

8. The method according to claim 1, characterized in that The first data features include supply statistical features and supply evaluation features, and the second data features include demand statistical features and demand evaluation features.

9. The method according to claim 1, characterized in that: The supply and demand forecast result includes a supply type and a demand type, wherein the supply type includes at least two supply grade types, and the demand type includes at least two grade types.

10. An information processing device, characterized in that: include: A supply information processing module, which obtains original description information of a supply object, determines supply expression data based on the original description information of the supply object, and determines a first data feature of the supply expression data; A demand information processing module, used to obtain original demand information, determine demand expression data based on the original demand information, and determine a second data feature of the demand expression data; The supply and demand forecasting module is used to perform supply and demand forecasting on the first data feature of the supply expression data and the second data feature of the demand expression data based on the supply and demand forecasting model to obtain a supply and demand forecasting result.

11. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can perform the information processing method according to any one of claims 1 to 9.

12. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the information processing method according to any one of claims 1 to 9 when executed.

Images