CN118014695B - Commodity pushing method and system based on multi-source data screening - Google Patents
Commodity pushing method and system based on multi-source data screening Download PDFInfo
- Publication number
- CN118014695B CN118014695B CN202410417618.1A CN202410417618A CN118014695B CN 118014695 B CN118014695 B CN 118014695B CN 202410417618 A CN202410417618 A CN 202410417618A CN 118014695 B CN118014695 B CN 118014695B
- Authority
- CN
- China
- Prior art keywords
- commodity
- searched
- data source
- data
- pushed
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 33
- 238000012216 screening Methods 0.000 title claims abstract description 24
- 230000006399 behavior Effects 0.000 claims abstract description 78
- 238000004364 calculation method Methods 0.000 claims abstract description 21
- 238000012790 confirmation Methods 0.000 claims description 8
- 238000001514 detection method Methods 0.000 claims description 5
- 238000010276 construction Methods 0.000 claims description 3
- 238000010586 diagram Methods 0.000 description 4
- 238000004891 communication Methods 0.000 description 3
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000001680 brushing effect Effects 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application discloses a commodity pushing method and a commodity pushing system based on multi-source data screening, which are used for collecting commodity total data of searched data sources, carrying out hot spot clustering on the commodity total data according to commodity caches of the searched data sources, and outputting a commodity total database; acquiring a commodity pushing request of a data source to be searched, and calculating a semantic distance from a commodity pushing result of the searched data source to obtain a potential pushed commodity; acquiring commodity cache of a data source to be searched, and carrying out semantic distance calculation on the commodity cache and scene information of commodity total data of the potential pushed commodity to obtain the potential pushed commodity; collecting user feedback behaviors of a data source to be searched, analyzing the user feedback behaviors according to commodity total data, and outputting commodity adjustment strategies fed back by a user; and reminding the data source to be searched and the commodity provider according to the commodity adjustment strategy. The method can reduce the request amount of the commodity full-quantity database, thereby reducing the machine occupation of the commodity full-quantity database and saving resources.
Description
Technical Field
The application relates to the technical field of big data, in particular to a commodity pushing method and system based on multi-source data screening.
Background
At present, an electronic commerce platform generally provides a display module of commodities needed by a user, a search function is provided, the user inputs commodity keywords to obtain a commodity list related to the commodity keywords, however, the sorting display of traditional commodities and searched results is generally performed according to the relevance, and the current sorting is often performed according to the commodity clicking times, payment methods and the like, so that the commodities of a plurality of merchants want to be sorted relatively forward, irregular bill brushing can be performed, people and sales are accumulated, the relevance is improved, the commodities and the searching sorting display of the current commerce platform are relatively disordered, the reliability is low, and the pressure of the user for screening information is increased.
The problem of information overload is increasingly serious nowadays, and the pertinence of an electronic commerce platform is not strong when commodity recommendation is carried out, so that the shopping experience of consumers is influenced.
Disclosure of Invention
The application mainly aims to provide a commodity pushing method and system based on multi-source data screening, which are used for solving the problem of inaccurate commodity pushing in the prior art.
In order to achieve the above purpose, the present application provides the following technical solutions:
According to a first aspect of the present invention, the present invention provides a commodity pushing method based on multi-source data screening, which is characterized by comprising:
acquiring commodity total data of a searched data source, performing hot spot clustering on the commodity total data according to a commodity cache of the searched data source, and outputting a commodity total database;
Acquiring commodity pushing requests of data sources to be searched, and calculating semantic distances between the commodity pushing requests and commodity pushing results of searched data sources in the commodity total database to obtain M potential pushed commodities with semantic distances smaller than a preset limit distance and ordered from small to large, wherein M is a natural number;
Acquiring commodity cache of the data source to be searched, and carrying out semantic distance calculation with scene information of commodity total data of the M potential pushed commodities to obtain potential pushed commodities with the minimum semantic distance as the commodity to be pushed of the data source to be searched;
Collecting user feedback behaviors of the data source to be searched, analyzing the user feedback behaviors according to the commodity total data of the commodity to be pushed, and outputting commodity adjustment strategies fed back by the user of the data source to be searched;
Reminding the data source to be searched and the commodity provider according to the commodity adjustment strategy, and feeding back the user feedback behavior of the adjusted data source to be searched to the commodity full database.
Further, the collecting the commodity total data of the searched data source, performing hot spot clustering on the commodity total data according to the commodity cache of the searched data source, outputting a commodity total database, and further comprising:
Acquiring commodity cache of the searched data source, and obtaining search time and search duration of the searched data source from the commodity cache;
acquiring commodity total data acquired when the searched data source performs user feedback detection during pushing selection;
performing hot spot clustering on the commodity total data according to the search time and the search duration of the searched data source, and outputting hot spot sub-clustering results under at least busy period, idle period and comprehensive period;
and adding labels to the hot spot sub-clustering results in the busy period, the idle period and the comprehensive period, associating the hot spot sub-clustering results with the corresponding searched data sources, and outputting a commodity total database.
Further, the acquiring a commodity pushing request of the data source to be searched, performing semantic distance calculation with a commodity pushing result of the searched data source in the commodity total database, and obtaining M potential pushed commodities with semantic distances smaller than a preset limit distance, wherein the M is a natural number, and the M potential pushed commodities are ordered from small to large according to the semantic distances, and the method further comprises the following steps:
Acquiring a commodity pushing request of a data source to be searched, and acquiring at least a searching position, searching time and searching frequency of the data source to be searched from the commodity pushing request;
acquiring commodity pushing results of searched data sources in the commodity total database, and at least acquiring search positions, search time and search frequency of the searched data sources from the commodity pushing results;
And assigning weights for the search position, the search time and the search frequency, obtaining the comprehensive semantic distance between the commodity pushing request of the data source to be searched and the commodity pushing result of the searched data source, and outputting M potential pushed commodities of which the comprehensive semantic distance is smaller than a preset limit distance and which are ordered from small to large according to the semantic distance.
Further, the acquiring the commodity cache of the data source to be searched, performing semantic distance calculation with the scene information of the commodity total data of the M potential pushed commodities, and obtaining the potential pushed commodity with the minimum semantic distance as the commodity to be pushed of the data source to be searched, further includes:
acquiring commodity cache of the data source to be searched, and acquiring intention keywords of the data source to be searched from the commodity cache of the data source to be searched;
Acquiring scene information of commodity total data of the M potential pushed commodities, and outputting scene confirmation results of the commodity total data of the potential pushed commodities;
And calculating the semantic distance between the intention keywords of the data source to be searched and the scene confirmation result to obtain a potential pushed commodity with the minimum semantic distance as the commodity to be pushed of the data source to be searched.
Further, the collecting the user feedback behavior of the data source to be searched, analyzing the user feedback behavior according to the commodity total data of the commodity to be pushed, and outputting a commodity adjustment strategy fed back by the user of the data source to be searched, and further comprises:
acquiring user feedback behaviors of the data source to be searched, dividing intervals of the user feedback behaviors, and outputting idle time behavior intervals and busy time behavior intervals of the user feedback behaviors;
acquiring commodity total data of the commodity to be pushed, and outputting idle time historical behavior intervals and busy time historical behavior intervals of the user feedback behaviors in the commodity total data;
Comparing the idle time period behavior interval and the busy time period behavior interval with the idle time period historical behavior interval and the busy time period historical behavior interval respectively, and outputting the deviation degree of the idle time period behavior interval and the busy time period behavior interval;
And adjusting the user feedback of the data source to be searched according to the deviation degree of the idle time interval and the busy time interval, and outputting a commodity adjustment strategy of the user feedback of the data source to be searched.
Further, the prompting is performed on the data source to be searched and the commodity provider according to the commodity adjustment strategy, and the feedback behavior of the user of the adjusted data source to be searched is fed back to the commodity full database, and the method further comprises:
When the commodity adjustment strategy is to increase the searching frequency of idle time searching of the data source to be searched, a first reminder is sent to the data source to be searched;
When the commodity adjustment strategy is to reduce the searching frequency of idle searching of the data source to be searched, a second reminder is sent to the data source to be searched and a commodity provider;
When the commodity adjustment strategy expands the range of a search engine for busy hour searching of the data source to be searched, a third reminder is sent to the data source to be searched;
When the commodity adjustment strategy is used for narrowing the range of a search engine for busy hour searching of the data source to be searched, a fourth reminder is sent to the data source to be searched;
and the data source to be searched carries out user feedback adjustment according to the prompt and/or the suggestion of the commodity provider, and the user feedback behavior of the adjusted data source to be searched is fed back to the commodity total database.
According to a second aspect of the present invention, the present invention claims a commodity pushing system based on multi-source data screening, which is characterized by comprising:
the historical library construction module is used for collecting commodity total data of the searched data source, carrying out hot spot clustering on the commodity total data according to the commodity cache of the searched data source, and outputting a commodity total database;
The candidate calculation module is used for collecting commodity pushing requests of data sources to be searched, carrying out semantic distance calculation on the commodity pushing requests and commodity pushing results of searched data sources in the commodity total database, and obtaining M potential pushed commodities with semantic distances smaller than a preset limit distance and ordered from small to large according to the semantic distances, wherein M is a natural number;
The commodity cache of the data source to be searched is collected, semantic distance calculation is carried out on the commodity cache of the data source to be searched and scene information of commodity total data of the M potential pushed commodities, and the potential pushed commodities with the minimum semantic distance are obtained to serve as the commodities to be pushed of the data source to be searched;
the adjustment module is used for acquiring user feedback behaviors of the data source to be searched, analyzing the user feedback behaviors according to the commodity total data of the commodity to be pushed and outputting commodity adjustment strategies fed back by the user of the data source to be searched;
the feedback module reminds the data source to be searched and the commodity provider according to the commodity adjustment strategy, and feeds back the feedback behavior of the user of the adjusted data source to be searched to the commodity total database;
The commodity pushing system based on the multi-source data screening is used for realizing the commodity pushing method based on the multi-source data screening.
The application discloses a commodity pushing method and a commodity pushing system based on multi-source data screening, which are used for collecting commodity total data of searched data sources, carrying out hot spot clustering on the commodity total data according to commodity caches of the searched data sources, and outputting a commodity total database; acquiring a commodity pushing request of a data source to be searched, and calculating a semantic distance from a commodity pushing result of the searched data source to obtain a potential pushed commodity; acquiring commodity cache of a data source to be searched, and carrying out semantic distance calculation on the commodity cache and scene information of commodity total data of the potential pushed commodity to obtain the potential pushed commodity; collecting user feedback behaviors of a data source to be searched, analyzing the user feedback behaviors according to commodity total data, and outputting commodity adjustment strategies fed back by a user; and reminding the data source to be searched and the commodity provider according to the commodity adjustment strategy. The method can reduce the request amount of the commodity full-quantity database, thereby reducing the machine occupation of the commodity full-quantity database and saving resources.
Drawings
FIG. 1 is a workflow diagram of a method for pushing merchandise based on multi-source data screening according to an embodiment of the present application;
FIG. 2 is a second workflow diagram of a method for pushing merchandise based on multi-source data screening according to an embodiment of the present application;
FIG. 3 is a third workflow diagram of a method for pushing merchandise based on multi-source data screening according to an embodiment of the present application;
FIG. 4 is a fourth flowchart illustrating a method for pushing commodities based on multi-source data screening according to an embodiment of the present application;
Fig. 5 is a structural block diagram of a commodity pushing system based on multi-source data screening according to an embodiment of the present application.
Detailed Description
The following description of the embodiments of the present application will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are only some, but not all embodiments of the application. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
The terms "first," "second," "third," and the like in this disclosure are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defining "first," "second," and "third" may explicitly or implicitly include at least one such feature. In the description of the present application, the meaning of "plurality" means at least two, for example, two, three, etc., unless specifically defined otherwise. All directional indications (such as up, down, left, right, front, back … …) in the embodiments of the present application are merely used to explain the relative positional relationship between the components, the exact search situation, etc. under a certain specific gesture (as shown in the drawings), and if the specific gesture is changed, the directional indication is changed accordingly. Furthermore, the terms "comprise" and "have," as well as any variations thereof, are intended to cover a non-exclusive inclusion. For example, a process, method, system, article, or apparatus that comprises a list of steps or elements is not limited to only those listed steps or elements but may include other steps or elements not listed or inherent to such process, method, article, or apparatus.
Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment may be included in at least one embodiment of the application. The appearances of such phrases in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. Those of skill in the art will explicitly and implicitly appreciate that the embodiments described herein may be combined with other embodiments.
According to a first embodiment of the present invention, referring to fig. 1, the present invention claims a commodity pushing method based on multi-source data screening, including:
acquiring commodity total data of a searched data source, performing hot spot clustering on the commodity total data according to a commodity cache of the searched data source, and outputting a commodity total database;
Acquiring commodity pushing requests of data sources to be searched, and calculating semantic distances between the commodity pushing requests and commodity pushing results of searched data sources in the commodity total database to obtain M potential pushed commodities with semantic distances smaller than a preset limit distance and ordered from small to large, wherein M is a natural number;
Acquiring commodity cache of the data source to be searched, and carrying out semantic distance calculation with scene information of commodity total data of the M potential pushed commodities to obtain potential pushed commodities with the minimum semantic distance as the commodity to be pushed of the data source to be searched;
Collecting user feedback behaviors of the data source to be searched, analyzing the user feedback behaviors according to the commodity total data of the commodity to be pushed, and outputting commodity adjustment strategies fed back by the user of the data source to be searched;
Reminding the data source to be searched and the commodity provider according to the commodity adjustment strategy, and feeding back the user feedback behavior of the adjusted data source to be searched to the commodity full database.
In this embodiment, the specific application environment is that in the commodity purchasing platform, there are many situations that the network communication faults cause that the time period cannot be searched accurately, and user feedback detection is performed on the data sources, so that the time period pushing situation of the data sources can be monitored well. In the scheme, the time period accurate searching and pushing process of accurate data sources which have completed searching in history is referred to, and compared with the current data sources which have not completed searching, the method and the device find the most conforming time period reference object to perform user feedback detection adjustment according to secondary semantic distance calculation.
Further, referring to fig. 2, the collecting the commodity total data of the searched data source, performing hot spot clustering on the commodity total data according to the commodity cache of the searched data source, outputting a commodity total database, and further includes:
Acquiring commodity cache of the searched data source, and obtaining search time and search duration of the searched data source from the commodity cache;
acquiring commodity total data acquired when the searched data source performs user feedback detection during pushing selection;
performing hot spot clustering on the commodity total data according to the search time and the search duration of the searched data source, and outputting hot spot sub-clustering results under at least busy period, idle period and comprehensive period;
and adding labels to the hot spot sub-clustering results in the busy period, the idle period and the comprehensive period, associating the hot spot sub-clustering results with the corresponding searched data sources, and outputting a commodity total database.
In this embodiment, the obtaining, in the commodity cache, the search time and the search duration of the searched data source at least includes:
Search time: morning, afternoon, evening, and early morning;
search duration: less than 10 minutes, 10-30 minutes, 30-90 minutes, greater than 90 minutes.
When the hot spot clustering is carried out on the commodity total data according to the searching time and the searching time of the searched data source, semantic analysis is carried out according to the searching time of the searched data source to obtain the hot spot sub-clustering result which can be confirmed to be the hot spot sub-clustering result in the idle period in the early morning; when the hot spot clustering can not be performed on the commodity total data according to the searching time of the searched data source, confirmation is performed according to the searching time, for example, the specific time period can not be judged in the afternoon, but the hot spot clustering result under the comprehensive time period is confirmed through the Saturday date.
Further, referring to fig. 3, the acquiring a commodity pushing request of a data source to be searched, performing semantic distance calculation with a commodity pushing result of a searched data source in the commodity total database, to obtain M potential pushed commodities with semantic distances smaller than a preset limit distance, wherein M is a natural number, and the M potential pushed commodities are ordered from small to large according to the semantic distances, and the method further includes:
Acquiring a commodity pushing request of a data source to be searched, and acquiring at least a searching position, searching time and searching frequency of the data source to be searched from the commodity pushing request;
acquiring commodity pushing results of searched data sources in the commodity total database, and at least acquiring search positions, search time and search frequency of the searched data sources from the commodity pushing results;
And assigning weights for the search position, the search time and the search frequency, obtaining the comprehensive semantic distance between the commodity pushing request of the data source to be searched and the commodity pushing result of the searched data source, and outputting M potential pushed commodities of which the comprehensive semantic distance is smaller than a preset limit distance and which are ordered from small to large according to the semantic distance.
In this embodiment, the search position, search time and search frequency of the data source to be searched and the search position, search time and search frequency of the searched data source are respectively compared; and assigning weights of 15%,35% and 50% to the search positions, the search time and the search frequency, calculating the numerical value of the search positions by adopting a cosine semantic distance calculation method, taking the cosine distances of the search time and the search frequency as comprehensive semantic distances, and outputting M potential pushed commodities which are smaller than a preset limit distance in the comprehensive semantic distances and are ordered from small to large in the semantic distances.
Further, referring to fig. 4, the collecting the commodity cache of the data source to be searched, performing semantic distance calculation with the scene information of the commodity total data of the M potential pushed commodities, to obtain the potential pushed commodity with the minimum semantic distance as the commodity to be pushed of the data source to be searched, further includes:
acquiring commodity cache of the data source to be searched, and acquiring intention keywords of the data source to be searched from the commodity cache of the data source to be searched;
Acquiring scene information of commodity total data of the M potential pushed commodities, and outputting scene confirmation results of the commodity total data of the potential pushed commodities;
And calculating the semantic distance between the intention keywords of the data source to be searched and the scene confirmation result to obtain a potential pushed commodity with the minimum semantic distance as the commodity to be pushed of the data source to be searched.
In this embodiment, after collecting M potential pushed commodities whose comprehensive semantic distance is smaller than a preset limit distance and which are ordered from small to large according to the semantic distance, a to-be-pushed commodity which best accords with the to-be-searched data source needs to be further found.
Obtaining an intention keyword of the data source to be searched from a commodity cache of the data source to be searched, wherein the intention keyword is specific to the details of a lower part of idle time searching or busy time searching;
and acquiring more specific intention keywords of the potential pushed commodities in busy time, idle time and comprehensive time when outputting a scene confirmation result of the commodity total data of the potential pushed commodities according to the scene information of the commodity total data of the M potential pushed commodities.
And carrying out cosine semantic distance based on the position values occupied by the intended keywords of the data source to be searched and the positions of the intended keywords of the potential pushed commodities, and selecting the potential pushed commodities with the minimum semantic distance as the commodities to be pushed of the data source to be searched.
Further, the collecting the user feedback behavior of the data source to be searched, analyzing the user feedback behavior according to the commodity total data of the commodity to be pushed, and outputting a commodity adjustment strategy fed back by the user of the data source to be searched, and further comprises:
acquiring user feedback behaviors of the data source to be searched, dividing intervals of the user feedback behaviors, and outputting idle time behavior intervals and busy time behavior intervals of the user feedback behaviors;
acquiring commodity total data of the commodity to be pushed, and outputting idle time historical behavior intervals and busy time historical behavior intervals of the user feedback behaviors in the commodity total data;
Comparing the idle time period behavior interval and the busy time period behavior interval with the idle time period historical behavior interval and the busy time period historical behavior interval respectively, and outputting the deviation degree of the idle time period behavior interval and the busy time period behavior interval;
And adjusting the user feedback of the data source to be searched according to the deviation degree of the idle time interval and the busy time interval, and outputting a commodity adjustment strategy of the user feedback of the data source to be searched.
In this embodiment, in order to better analyze the feedback behavior of the user, the intervals are divided, and then the positions of the intervals are compared in more detail, so that the area ratios in different intervals are collected as the deviation degree of the behavior intervals.
Further, the prompting is performed on the data source to be searched and the commodity provider according to the commodity adjustment strategy, and the feedback behavior of the user of the adjusted data source to be searched is fed back to the commodity full database, and the method further comprises:
When the commodity adjustment strategy is to increase the searching frequency of idle time searching of the data source to be searched, a first reminder is sent to the data source to be searched;
When the commodity adjustment strategy is to reduce the searching frequency of idle searching of the data source to be searched, a second reminder is sent to the data source to be searched and a commodity provider;
When the commodity adjustment strategy expands the range of a search engine for busy hour searching of the data source to be searched, a third reminder is sent to the data source to be searched;
When the commodity adjustment strategy is used for narrowing the range of a search engine for busy hour searching of the data source to be searched, a fourth reminder is sent to the data source to be searched;
and the data source to be searched carries out user feedback adjustment according to the prompt and/or the suggestion of the commodity provider, and the user feedback behavior of the adjusted data source to be searched is fed back to the commodity total database.
In this embodiment, when the commodity adjustment policy is to increase the searching frequency for idle searching of the to-be-searched data source or expand the range of a search engine for busy searching of the to-be-searched data source, it indicates that the update rate of the to-be-searched data source relative to the to-be-updated commodity is too slow, and at this time, only the to-be-searched data source needs to be reminded to confirm and determine whether to increase the update speed;
And when the commodity adjustment policy is to reduce the searching frequency of idle searching of the data source to be searched or the commodity adjustment policy is to narrow the searching engine range of busy searching of the data source to be searched, the commodity adjustment policy indicates that the updating speed of the data source to be searched relative to the commodity to be updated is too fast, and at the moment, the personal of the data source to be searched and the commodity provider should be reminded to confirm whether the updating speed needs to be slowed down or not.
According to a second embodiment of the present invention, referring to fig. 5, the present invention claims a merchandise pushing system based on multi-source data screening, comprising:
the historical library construction module is used for collecting commodity total data of the searched data source, carrying out hot spot clustering on the commodity total data according to the commodity cache of the searched data source, and outputting a commodity total database;
The candidate calculation module is used for collecting commodity pushing requests of data sources to be searched, carrying out semantic distance calculation on the commodity pushing requests and commodity pushing results of searched data sources in the commodity total database, and obtaining M potential pushed commodities with semantic distances smaller than a preset limit distance and ordered from small to large according to the semantic distances, wherein M is a natural number;
The commodity cache of the data source to be searched is collected, semantic distance calculation is carried out on the commodity cache of the data source to be searched and scene information of commodity total data of the M potential pushed commodities, and the potential pushed commodities with the minimum semantic distance are obtained to serve as the commodities to be pushed of the data source to be searched;
the adjustment module is used for acquiring user feedback behaviors of the data source to be searched, analyzing the user feedback behaviors according to the commodity total data of the commodity to be pushed and outputting commodity adjustment strategies fed back by the user of the data source to be searched;
the feedback module reminds the data source to be searched and the commodity provider according to the commodity adjustment strategy, and feeds back the feedback behavior of the user of the adjusted data source to be searched to the commodity total database;
The commodity pushing system based on the multi-source data screening is used for realizing the commodity pushing method based on the multi-source data screening.
In the several embodiments provided in the present application, it should be understood that the disclosed system, apparatus and method may be implemented in other manners. For example, the apparatus embodiments described above are merely illustrative, e.g., the division of elements is merely a logical functional division, and there may be additional divisions of actual implementation, e.g., multiple elements or components may be combined or integrated into another system, or some features may be omitted, or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be an indirect coupling or communication connection via some interfaces, devices or units, which may be in electrical, mechanical or other forms.
In addition, each functional unit in the embodiments of the present application may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units may be implemented in hardware or in software functional units. The foregoing is only the embodiments of the present application, and the patent scope of the application is not limited thereto, but is also covered by the patent protection scope of the application, as long as the equivalent structure or equivalent flow changes made by the description and the drawings of the application or the direct or indirect application in other related technical fields are adopted.
The embodiments of the application have been described in detail above, but they are merely examples, and the application is not limited to the above-described embodiments. It will be apparent to those skilled in the art that any equivalent modifications or substitutions to this application are within the scope of the application, and therefore, all equivalent changes and modifications, improvements, etc. that do not depart from the spirit and scope of the principles of the application are intended to be covered by this application.
Claims (2)
1. The commodity pushing method based on multi-source data screening is characterized by comprising the following steps of:
acquiring commodity total data of a searched data source, performing hot spot clustering on the commodity total data according to a commodity cache of the searched data source, and outputting a commodity total database;
Acquiring commodity pushing requests of data sources to be searched, and calculating semantic distances between the commodity pushing requests and commodity pushing results of searched data sources in the commodity total database to obtain M potential pushed commodities with semantic distances smaller than a preset limit distance and ordered from small to large, wherein M is a natural number;
Acquiring commodity cache of the data source to be searched, and carrying out semantic distance calculation with scene information of commodity total data of the M potential pushed commodities to obtain potential pushed commodities with the minimum semantic distance as the commodity to be pushed of the data source to be searched;
Collecting user feedback behaviors of the data source to be searched, analyzing the user feedback behaviors according to the commodity total data of the commodity to be pushed, and outputting commodity adjustment strategies fed back by the user of the data source to be searched;
Reminding the data source to be searched and the commodity provider according to the commodity adjustment strategy, and feeding back the user feedback behavior of the adjusted data source to be searched to the commodity full database;
the method comprises the steps of collecting commodity full data of a searched data source, carrying out hot spot clustering on the commodity full data according to commodity cache of the searched data source, outputting a commodity full database, and further comprising:
Acquiring commodity cache of the searched data source, and obtaining search time and search duration of the searched data source from the commodity cache;
acquiring commodity total data acquired when the searched data source performs user feedback detection during pushing selection;
performing hot spot clustering on the commodity total data according to the search time and the search duration of the searched data source, and outputting hot spot sub-clustering results under at least busy period, idle period and comprehensive period;
Adding labels to the hot spot sub-clustering results in the busy period, the idle period and the comprehensive period, associating the hot spot sub-clustering results with the corresponding searched data sources, and outputting a commodity total database;
The method comprises the steps of acquiring a commodity pushing request of a data source to be searched, calculating semantic distances with commodity pushing results of searched data sources in a commodity total database, and obtaining M potential pushed commodities with semantic distances smaller than a preset limit distance and ordered from small to large according to the semantic distances, wherein M is a natural number, and the method further comprises the steps of:
Acquiring a commodity pushing request of a data source to be searched, and acquiring at least a searching position, searching time and searching frequency of the data source to be searched from the commodity pushing request;
acquiring commodity pushing results of searched data sources in the commodity total database, and at least acquiring search positions, search time and search frequency of the searched data sources from the commodity pushing results;
Assigning weights for the search position, the search time and the search frequency, obtaining comprehensive semantic distances between the commodity pushing request of the data source to be searched and commodity pushing results of the searched data source, and outputting M potential pushed commodities, wherein the comprehensive semantic distances are smaller than a preset limit distance and are ordered from small to large according to semantic distances;
The acquiring the commodity cache of the data source to be searched, performing semantic distance calculation with the scene information of the commodity total data of the M potential pushed commodities to obtain the potential pushed commodity with the minimum semantic distance as the commodity to be pushed of the data source to be searched, and further comprising:
acquiring commodity cache of the data source to be searched, and acquiring intention keywords of the data source to be searched from the commodity cache of the data source to be searched;
Acquiring scene information of commodity total data of the M potential pushed commodities, and outputting scene confirmation results of the commodity total data of the potential pushed commodities;
calculating semantic distances between the intention keywords of the data source to be searched and the scene confirmation result to obtain potential pushed commodities with the minimum semantic distances as the commodities to be pushed of the data source to be searched;
the step of collecting the user feedback behavior of the data source to be searched, analyzing the user feedback behavior according to the commodity total data of the commodity to be pushed, and outputting a commodity adjustment strategy fed back by the user of the data source to be searched, and further comprises the following steps:
acquiring user feedback behaviors of the data source to be searched, dividing intervals of the user feedback behaviors, and outputting idle time behavior intervals and busy time behavior intervals of the user feedback behaviors;
acquiring commodity total data of the commodity to be pushed, and outputting idle time historical behavior intervals and busy time historical behavior intervals of the user feedback behaviors in the commodity total data;
Comparing the idle time period behavior interval and the busy time period behavior interval with the idle time period historical behavior interval and the busy time period historical behavior interval respectively, and outputting the deviation degree of the idle time period behavior interval and the busy time period behavior interval;
Adjusting the user feedback of the data source to be searched according to the deviation degree of the idle time interval and the busy time interval, and outputting a commodity adjustment strategy of the user feedback of the data source to be searched;
Reminding the data source to be searched and the commodity provider according to the commodity adjustment strategy, and feeding back the user feedback behavior of the adjusted data source to be searched to the commodity full database, and further comprising:
When the commodity adjustment strategy is to increase the searching frequency of idle time searching of the data source to be searched, a first reminder is sent to the data source to be searched;
When the commodity adjustment strategy is to reduce the searching frequency of idle searching of the data source to be searched, a second reminder is sent to the data source to be searched and a commodity provider;
When the commodity adjustment strategy expands the range of a search engine for busy hour searching of the data source to be searched, a third reminder is sent to the data source to be searched;
When the commodity adjustment strategy is used for narrowing the range of a search engine for busy hour searching of the data source to be searched, a fourth reminder is sent to the data source to be searched;
and the data source to be searched carries out user feedback adjustment according to the prompt and/or the suggestion of the commodity provider, and the user feedback behavior of the adjusted data source to be searched is fed back to the commodity total database.
2. A merchandise pushing system based on multi-source data screening, comprising:
the historical library construction module is used for collecting commodity total data of the searched data source, carrying out hot spot clustering on the commodity total data according to the commodity cache of the searched data source, and outputting a commodity total database;
The candidate calculation module is used for collecting commodity pushing requests of data sources to be searched, carrying out semantic distance calculation on the commodity pushing requests and commodity pushing results of searched data sources in the commodity total database, and obtaining M potential pushed commodities with semantic distances smaller than a preset limit distance and ordered from small to large according to the semantic distances, wherein M is a natural number;
The commodity cache of the data source to be searched is collected, semantic distance calculation is carried out on the commodity cache of the data source to be searched and scene information of commodity total data of the M potential pushed commodities, and the potential pushed commodities with the minimum semantic distance are obtained to serve as the commodities to be pushed of the data source to be searched;
the adjustment module is used for acquiring user feedback behaviors of the data source to be searched, analyzing the user feedback behaviors according to the commodity total data of the commodity to be pushed and outputting commodity adjustment strategies fed back by the user of the data source to be searched;
the feedback module reminds the data source to be searched and the commodity provider according to the commodity adjustment strategy, and feeds back the feedback behavior of the user of the adjusted data source to be searched to the commodity total database;
The commodity pushing system based on multi-source data screening is used for realizing the commodity pushing method based on multi-source data screening as set forth in claim 1.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410417618.1A CN118014695B (en) | 2024-04-09 | 2024-04-09 | Commodity pushing method and system based on multi-source data screening |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410417618.1A CN118014695B (en) | 2024-04-09 | 2024-04-09 | Commodity pushing method and system based on multi-source data screening |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118014695A CN118014695A (en) | 2024-05-10 |
| CN118014695B true CN118014695B (en) | 2024-06-04 |
Family
ID=90948811
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410417618.1A Active CN118014695B (en) | 2024-04-09 | 2024-04-09 | Commodity pushing method and system based on multi-source data screening |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118014695B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111814058A (en) * | 2020-08-20 | 2020-10-23 | 深圳市欢太科技有限公司 | Pushing method and device based on user intention, electronic equipment and storage medium |
| CN116861060A (en) * | 2023-07-04 | 2023-10-10 | 上海微盟企业发展有限公司 | Private domain electronic commerce data searching method, device, equipment and storage medium |
| CN117237038A (en) * | 2023-08-22 | 2023-12-15 | 雨果跨境(厦门)科技有限公司 | Commodity accurate exposure processing system based on flow engine |
| CN117372128A (en) * | 2023-10-26 | 2024-01-09 | 广东德澳智慧医疗科技有限公司 | Meta universe virtual shopping center innovation experience platform serving QoS |
-
2024
- 2024-04-09 CN CN202410417618.1A patent/CN118014695B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111814058A (en) * | 2020-08-20 | 2020-10-23 | 深圳市欢太科技有限公司 | Pushing method and device based on user intention, electronic equipment and storage medium |
| CN116861060A (en) * | 2023-07-04 | 2023-10-10 | 上海微盟企业发展有限公司 | Private domain electronic commerce data searching method, device, equipment and storage medium |
| CN117237038A (en) * | 2023-08-22 | 2023-12-15 | 雨果跨境(厦门)科技有限公司 | Commodity accurate exposure processing system based on flow engine |
| CN117372128A (en) * | 2023-10-26 | 2024-01-09 | 广东德澳智慧医疗科技有限公司 | Meta universe virtual shopping center innovation experience platform serving QoS |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118014695A (en) | 2024-05-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US12001496B1 (en) | Selecting, ranking, and/or presenting microsite content | |
| US7620651B2 (en) | System for dynamic product summary based on consumer-contributed keywords | |
| CN109840782B (en) | Click rate prediction method, device, server and storage medium | |
| AU2022201654A1 (en) | System and engine for seeded clustering of news events | |
| KR100492075B1 (en) | Method and system for selecting search list table in an internet search engine in response to search request | |
| KR100469824B1 (en) | Method and system for managing the impressing of the search listing based on advertisement group | |
| US8799275B2 (en) | Information retrieval based on semantic patterns of queries | |
| US10348550B2 (en) | Method and system for processing network media information | |
| Aslam et al. | TREC 2013 Temporal Summarization. | |
| US7831474B2 (en) | System and method for associating an unvalued search term with a valued search term | |
| CN112052394B (en) | Professional content information recommendation method, system, terminal equipment and storage medium | |
| CN102193929A (en) | Method and equipment for determining word information entropy and searching by using word information entropy | |
| CN111028087B (en) | Information display method, device and equipment | |
| CN110111167A (en) | A kind of method and apparatus of determining recommended | |
| CN106126544B (en) | Internet content delivery method and device | |
| CN105069036A (en) | Information recommendation method and apparatus | |
| US11941073B2 (en) | Generating and implementing keyword clusters | |
| CN111191111A (en) | Content recommendation method, device and storage medium | |
| CN112699309A (en) | Resource recommendation method, device, readable medium and equipment | |
| CN110880127A (en) | Consumption level prediction method and device, electronic equipment and storage medium | |
| CN118014695B (en) | Commodity pushing method and system based on multi-source data screening | |
| KR20050123236A (en) | Method and system for selecting search list table in an internet search engine in response to search request | |
| US20130185315A1 (en) | Identification of Events of Interest | |
| KR101172906B1 (en) | Method and system for managing the impressing of the search listing based on advertisement group | |
| CN115146169A (en) | Conversion rate model training method and device, electronic 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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |