Recommendation method, device, equipment and storage medium based on weather search

Abstract

The present invention discloses a recommendation method, device, equipment and storage medium based on weather search, the method comprising: upon receiving a weather search request for a target location from a user within a high-frequency search period corresponding to a weather search demand, obtaining weather information corresponding to the target location, wherein the high-frequency period corresponding to the weather search demand refers to a period of time within a day when the user traffic for weather search is greater than a preset traffic; obtaining a target portrait of the user, and determining target recommendation content based on the target portrait; and pushing the weather information corresponding to the target location and the target recommendation content to the user. In the above scheme, the target recommendation content determined based on the target portrait will be more in line with the user's search preferences and user needs, and personalized content recommendations will be made to the user while meeting the user's weather search needs, thereby increasing the diversity of recommendations.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a weather search-based recommendation method, device, equipment and storage medium. Background Art

[0002] With the continuous development of science and technology, Internet technology has also been rapidly improved. Users can search for content of interest through the Internet, among which weather search is one of them. As a high-frequency search service, weather search can quickly respond to user search operations and present weather information to users.

[0003] In the prior art, when a user searches for weather, the corresponding weather forecast, life index, meteorological index, etc. are usually pushed to the user, and the pushed content is relatively simple. Summary of the invention

[0004] The embodiments of the present invention solve the technical problem that the content of weather search-based recommendation is relatively single by providing a weather search-based recommendation method, device, equipment and storage medium.

[0005] In a first aspect, an embodiment of the present invention provides a weather search recommendation method, comprising:

[0006] When a user's weather search request for a target location is received within a high-frequency search period corresponding to a weather search demand, weather information corresponding to the target location is obtained, wherein the high-frequency period corresponding to the weather search demand refers to a period of time in a day when the user flow for weather search is greater than a preset flow;

[0007] Obtaining a target profile of the user, and determining target recommended content based on the target profile;

[0008] The weather information corresponding to the target location and the target recommended content are pushed to the user.

[0009] Optionally, determining target recommended content based on the target portrait includes:

[0010] Acquire an initial recommended content set, and determine attribute information of each recommended content in the initial recommended content set;

[0011] Based on the target portrait and the attribute information of each recommended content, scoring the relevance of each recommended content to the target portrait;

[0012] A preset number of recommended contents with larger scores are selected as the target recommended contents.

[0013] Optionally, the attribute information includes at least one of the following attributes: a category label corresponding to the recommended content, a historical click rate of the recommended content, a publishing source of the recommended content, a publishing time of the recommended content, and a content value score of the recommended content.

[0014] Optionally, the pushing of weather information corresponding to the target location and the target recommended content to the user includes:

[0015] Constructing a target push interface, wherein the target push interface includes a first area and a second area;

[0016] Fill the weather information in the first area, and fill the brief information of the target recommended content in the second area;

[0017] The target push interface is pushed to the user so that the user's terminal device displays the target push interface.

[0018] Optionally, a jump entry is further provided in the second area, and the method further includes:

[0019] When a target operation for the jump entry is received, a secondary display interface is pushed to the user; wherein the secondary display interface is used to display other recommended content, and the other recommended content is different from the target recommended content displayed in the second area.

[0020] Optionally, the method further includes:

[0021] When the weather search request is received within a non-high-frequency search period corresponding to the weather search demand, obtaining weather information corresponding to the target location, wherein the non-high-frequency period corresponding to the weather search demand refers to a period of time in a day when the user traffic for weather search is less than or equal to the preset traffic;

[0022] A weather display interface is pushed to the user, wherein the weather display interface is used to display the weather information.

[0023] Optionally, obtaining the initial recommended content set includes:

[0024] The initial recommended content set is screened out from a preset recommended content library based on the attribute information of the user.

[0025] Optionally, the filtering out the initial recommended content set from a preset recommended content library based on the attribute information of the user includes:

[0026] determining the user's usual place of residence;

[0027] If the target location is different from the permanent location, selecting recommended content corresponding to the target location from the preset recommended content library as the initial recommended content set;

[0028] If the target location is the same as the permanent location, the recommended content corresponding to the permanent location is selected from the preset recommended content library as the initial recommended content set.

[0029] Optionally, the filtering out the initial recommended content set from a preset recommended content library based on the attribute information of the user includes:

[0030] Determine the user's historical high-frequency search terms, where the historical high-frequency search terms are keywords that the user searched for with a frequency greater than a preset frequency within a preset time period before and / or after searching for weather in a historical period;

[0031] The recommended content corresponding to the historical high-frequency search terms is screened out from the preset recommended content library as the initial recommended content set.

[0032] Optionally, the filtering out the initial recommended content set from a preset recommended content library based on the attribute information of the user includes:

[0033] Determine a target group label of a target user group to which the user belongs;

[0034] The recommended content corresponding to the target group label is screened out from the preset recommended content library as the initial recommended content set.

[0035] Optionally, obtaining a target portrait corresponding to the user includes:

[0036] Acquire a target identifier corresponding to the user, where the target identifier includes at least one of a user identifier and a device identifier;

[0037] Based on a preset correspondence between the identifier and the user portrait, a target user portrait corresponding to the target identifier is determined as the target portrait.

[0038] In a second aspect, an embodiment of the present invention provides a recommendation device based on weather search, comprising:

[0039] A weather information processing module, configured to obtain weather information corresponding to a target location when receiving a weather search request from a user within a high-frequency search period corresponding to a weather search demand, wherein the high-frequency period corresponding to the weather search demand refers to a period of time in a day when the user flow for weather search is greater than a preset flow;

[0040] A recommended content processing module, used to obtain a target profile of the user and determine target recommended content based on the target profile;

[0041] The push module is used to push the weather information corresponding to the target location and the target recommended content to the user.

[0042] Optionally, the recommended content processing module is used to:

[0043] Acquire an initial recommended content set, and determine attribute information of each recommended content in the initial recommended content set;

[0044] Based on the target portrait and the attribute information of each recommended content, scoring the relevance of each recommended content to the target portrait;

[0045] A preset number of recommended contents with larger scores are selected as the target recommended contents.

[0046] Optionally, the attribute information includes at least one of the following attributes: a category label corresponding to the recommended content, a historical click rate of the recommended content, a publishing source of the recommended content, a publishing time of the recommended content, and a content value score of the recommended content.

[0047] Optionally, the push module is used to:

[0048] Constructing a target push interface, wherein the target push interface includes a first area and a second area;

[0049] Fill the weather information in the first area, and fill the brief information of the target recommended content in the second area;

[0050] The target push interface is pushed to the user so that the user's terminal device displays the target push interface.

[0051] Optionally, a jump entrance is further provided in the second area, and the device further comprises:

[0052] A secondary interface push module is used to push a secondary display interface to the user when receiving a target operation for the jump entry; wherein the secondary display interface is used to display other recommended content, and the other recommended content is different from the target recommended content displayed in the second area.

[0053] Optionally, the device further comprises:

[0054] an acquisition module, configured to acquire weather information corresponding to the target location upon receiving the weather search request within a non-high-frequency search period corresponding to the weather search demand, wherein the non-high-frequency period corresponding to the weather search demand refers to a period of time within a day when the user traffic for weather search is less than or equal to the preset traffic;

[0055] A weather push module is used to push a weather display interface to the user, wherein the weather display interface is used to display the weather information.

[0056] Optionally, the recommended content processing module is used to:

[0057] The initial recommended content set is screened out from a preset recommended content library based on the attribute information of the user.

[0058] Optionally, the recommended content processing module is used to:

[0059] determining the user's usual place of residence;

[0060] If the target location is different from the permanent location, selecting recommended content corresponding to the target location from the preset recommended content library as the initial recommended content set;

[0061] If the target location is the same as the permanent location, the recommended content corresponding to the permanent location is selected from the preset recommended content library as the initial recommended content set.

[0062] Optionally, the recommended content processing module is used to:

[0063] Determine the user's historical high-frequency search terms, where the historical high-frequency search terms are keywords that the user searched for with a frequency greater than a preset frequency within a preset time period before and / or after searching for weather in a historical period;

[0064] The recommended content corresponding to the historical high-frequency search terms is screened out from the preset recommended content library as the initial recommended content set.

[0065] Optionally, the recommended content processing module is used to:

[0066] Determine a target group label of a target user group to which the user belongs;

[0067] The recommended content corresponding to the target group label is screened out from the preset recommended content library as the initial recommended content set.

[0068] Optionally, the recommended content processing module is used to:

[0069] Acquire a target identifier corresponding to the user, where the target identifier includes at least one of a user identifier and a device identifier;

[0070] Based on a preset correspondence between the identifier and the user portrait, a target user portrait corresponding to the target identifier is determined as the target portrait.

[0071] In a third aspect, an embodiment of the present invention provides an electronic device comprising a memory and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by one or more processors to execute operation instructions corresponding to the weather search-based recommendation method provided in the first aspect.

[0072] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps corresponding to the weather search-based recommendation method provided in the first aspect.

[0073] One or more technical solutions provided by the embodiments of the present invention achieve at least the following technical effects or advantages:

[0074] The solution provided by the embodiment of the present invention, after receiving a weather search request from a user for a target location during a high-frequency search period corresponding to the weather search demand, obtains weather information corresponding to the weather search request on the one hand, and obtains a target portrait corresponding to the user on the other hand, and determines target recommendation content based on the target portrait. Since the target portrait can reflect the user's search habits and preferences, the target recommendation content determined based on the target portrait will be more in line with the user's search preferences and user needs. Pushing weather information and target recommendation content to the user, while satisfying the user's weather search needs, makes personalized content recommendations to the user, thereby improving the diversity of recommendations. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0076] Figure 1 A flow chart of a weather search-based recommendation method provided by an embodiment of the present invention;

[0077] Figure 2 A schematic diagram of a push interface provided by an embodiment of the present invention;

[0078] Figure 3A schematic diagram of a weather search-based recommendation device provided by an embodiment of the present invention;

[0079] Figure 4 A schematic diagram of an electronic device provided by an embodiment of the present invention;

[0080] Figure 5 A schematic diagram of the structure of a server provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0081] The present invention provides a weather search-based recommendation method, device, equipment and storage medium to solve the technical problem that the content of weather search-based recommendation is relatively single. The overall idea is as follows:

[0082] When a weather search request for a target location is received from a user during a high-frequency search period corresponding to the weather search demand, weather information corresponding to the target location is obtained, wherein the high-frequency period corresponding to the weather search demand refers to a time period in a day when the user traffic for weather search is greater than a preset traffic; the user's target portrait is obtained, and target recommendation content is determined based on the target portrait; the weather information and target recommendation content corresponding to the target location are pushed to the user.

[0083] Through the above technical solution, when a weather search request initiated by a user is received during a high-frequency period corresponding to the weather search demand, the weather information corresponding to the weather search request is obtained on the one hand, and the target portrait corresponding to the user is obtained on the other hand, and the target recommendation content is determined based on the target portrait. Since the target portrait can reflect the user's search habits and preferences, the target recommendation content determined based on the target portrait will be more in line with the user's search preferences and user needs. Pushing weather information and target recommendation content to users, while satisfying the user's weather search needs, provides personalized content recommendations to users, thereby improving the diversity of recommendations.

[0084] Please refer to Figure 1 , is a flow chart of a weather search-based recommendation method provided by an embodiment of the present invention, the method comprising the following steps:

[0085] Step S101: when a user's weather search request for a target location is received within a high-frequency search period corresponding to a weather search demand, weather information corresponding to the target location is obtained, wherein the high-frequency search period corresponding to the weather search demand refers to a period of time in a day when the user flow for weather search is greater than a preset flow;

[0086] Step S102: Obtain the user's target profile, and determine target recommendation content based on the target profile;

[0087] Step S103: Pushing weather information corresponding to the target location and the target recommended content to the user.

[0088] The method provided in the embodiment of the present invention can be applied to terminal devices, such as smart phones, tablet computers, etc., can also be applied to servers that establish data interaction with terminal devices, and can also be applied to systems composed of terminal devices and servers, without limitation here.

[0089] In step S101, the user's terminal device may run an application that provides weather search services, such as a search engine, a weather search application, a weather search applet, etc. The user initiates a weather search request by operating these applications. Specifically, when the user uses the terminal device, if the user searches for weather through the search engine of the terminal device, the user can enter the weather search keyword in the search box provided by the search engine.

[0090] Among them, the search request for weather can be a display request, that is, the request directly includes the weather search demand, for example, the user enters "Beijing weather" or directly enters "weather". If the user enters the location he wants to query, the location entered by the user is used as the target location. If the user does not enter the location for the weather search, it is assumed that the user is querying the weather at the user's location, and the user's location can be used as the target location. If the user uses the weather search application to search for weather, the user can manage the weather in multiple locations. For example, if the user adds multiple locations to the weather search application, the user's selection operation for the location can be detected. If the user selects location A, location A is used as the target location, and a weather search request for location A is initiated.

[0091] Among them, the search request for weather can also be an implicit request, that is, the request does not directly contain the demand for weather search, but through historical search data, it can be determined that the user's search request implicitly contains the demand for weather search. For example, if a user enters "Beijing" or "Zhangjiajie Scenic Area", based on historical search experience, it can be considered that the user has the possibility of wanting to obtain the weather conditions of the corresponding location, and the request can be considered a weather search request.

[0092] In the embodiment of the present invention, the high-frequency search period corresponding to the weather search demand can be pre-set, or can be obtained by statistically analyzing the historical user traffic corresponding to the weather search. For example, if the statistical results show that the user traffic for weather search in the period of 5:30-7:30 in the morning is greater than the preset traffic, then 5:30-7:30 in the morning is used as the high-frequency search period, wherein the preset traffic can be set according to actual needs and is not limited here.

[0093] If a weather search request input by a user is received during a high-frequency search period corresponding to a weather search demand, the weather search request is responded to and the weather information of the target location included in the request is determined. The weather information may include temperature information, air quality information, wind speed and direction information, etc. for the past week or the past 15 days. In addition, considering that during the high-frequency search period corresponding to the weather search demand, in addition to performing weather searches, users may also query and browse other content of interest. In order to reduce the user's query cost and realize the rapid push of content of interest to the user, in the embodiment of the specification of this application, the user's target portrait can be obtained through step S102, and based on the target portrait, the target recommended content that meets the user's preferences can be determined.

[0094] User portraits, that is, the structuring and labeling of user information, accurately depict and analyze all aspects of user information by describing the data of various dimensions of the user (such as age, gender, interest preferences, etc.). Therefore, based on user portraits, personalized recommendations can be made accurately.

[0095] Determining target recommended content based on target portrait can be achieved in many ways, for example, by calculating the similarity between each recommended content and the user portrait, and taking the recommended content with a similarity greater than a threshold as the target recommended content. Alternatively, recommendations can be made based on the target location of the user's weather search and the user portrait. If the target location is different from the user's permanent residence, then the user may go to the target location for fun or business trips, so the recommended content corresponding to the target location can be screened out, and then the user portrait is combined to determine the target recommended content that the user is interested in.

[0096] In the embodiment of the present specification, in order to accurately determine the target recommended content, step S102 can be implemented in the following manner: obtain an initial recommended content set, and determine the attribute information of each recommended content in the initial recommended content set; based on the target portrait of the user and the attribute information of each recommended content, score the degree of relevance of each recommended content to the target portrait; select a preset number of recommended content with larger scores as the target recommended content.

[0097] The initial recommended content set may be all data that can be used as recommended content. For example, when searching for weather through a search engine, the initial recommended content set may be recommended content in all fields that the search engine can obtain, including articles, videos, audio, pictures, etc. The initial recommended content set may also be filtered data, for example, all data obtained by the search engine is filtered based on the user profile. For example, if the user profile shows that the user is more interested in financial and sports content, then the recommended content of the financial and sports categories is selected as the preset recommended content set.

[0098] The attributes of the recommended content include but are not limited to at least one of the following attributes: the category label corresponding to the recommended content, the historical click rate of the recommended content, the publishing source of the recommended content, the publishing time of the recommended content, and the content value score of the recommended content.

[0099] Among them, the category label corresponding to the recommended content is used to represent the category to which the recommended content belongs. For example, the category label may include "finance", "health", "fitness", "beauty", etc.

[0100] The historical click-through rate of recommended content refers to the probability that a user clicks on the recommended content when all content under the category label of the recommended content is pushed to the user during the historical push process. For example, if the category label of a recommended content is "sports", the historical click-through rate of the recommended content refers to the probability that a user clicks on the recommended content when all content under the "sports" category is pushed to the user in the historical data.

[0101] The source of recommended content refers to the information of the publishing organization, publishing website, author, etc. For example, if the publishing website of a recommended content is Eastmoney.com, then the source of the recommended content is "Eastmoney.com".

[0102] The release time of recommended content refers to the time when the recommended content is first published or posted.

[0103] The content value score of the recommended content is used to characterize the quality of the recommended content. For example, if the recommended content contains rich information, little invalid content, and a low text repetition rate, then the corresponding content quality is good and the content value score is also high; if the recommended content contains little information, the content is seriously inconsistent with the title, there is a lot of invalid content, and the text repetition rate is high, then the corresponding content quality is poor and the content value score is also low.

[0104] In the embodiments of this specification, after determining the attribute information of each recommended content, the relevance of each recommended content to the target profile is scored based on the attribute information of the recommended content, with high scores for high relevance and low scores for low relevance. In the specific implementation process, the scoring of recommended content can be achieved in a variety of ways, such as but not limited to: score = (relevance + click-through rate) × authority × timeliness × content value, or score = (relevance + click-through rate) × authority × timeliness, etc.

[0105] Specifically, relevance refers to the correlation between the category label of the recommended content and the target portrait; click-through rate refers to the historical click-through rate of the recommended content; authority refers to the authority of the source of the recommended content; timeliness refers to the timeliness evaluation value corresponding to the release time of the recommended content; and content value refers to the content value score of the recommended content.

[0106] Among them, for authority, the authority of the publishing sources involved in the recommended content can be evaluated. For example, for recommended financial content, large-scale websites such as Dongfang Fortune Network and Tonghuashun Fortune Network have higher authority, while some unknown small financial websites have lower authority. Authority can be specifically quantified as a numerical value. The larger the numerical value, the higher the authority, and the smaller the numerical value, the lower the authority.

[0107] Timeliness can be determined based on the release time of the recommended content. The closer the release time is to the current weather search time, the higher the timeliness. The farther the release time is from the current weather search time, the lower the timeliness. Timeliness can also be quantified as a numerical value. The larger the numerical value, the better the timeliness, and the closer the release time is to the current weather search time. The smaller the numerical value, the worse the timeliness, and the farther the release time is from the current weather search time.

[0108] Through the above scoring formula, each recommended content can be scored, and the score can accurately represent the degree of association between the recommended content and the user profile and the reliability of the recommended content. Therefore, when determining the target recommended content to be pushed, the recommended content can be sorted from large to small by score, and the preset number of recommended content ranked in front is used as the target recommended content. The preset number can be set according to actual needs and is not limited here.

[0109] Of course, the scoring formula for the recommended content provided above is only one implementation method provided in the embodiments of this specification. Those skilled in the art can change the parameters and calculation methods of the above formula as needed, as long as the relevance between the recommended content and the user can be determined, and there is no limitation here.

[0110] In addition, when scoring the recommended content, it is also possible to consider whether the target location is a remote location or a permanent location. For example, if the target location is a remote location, the weight of the recommended content of the target location can be increased, and the weight of the recommended content of the permanent location can be reduced. In this way, when searching for weather in a remote location, the recommended content of the remote location can be recommended to the user first, so as to meet the user's travel needs in a remote location as much as possible.

[0111] After determining the weather information and target recommended content, the weather information and target recommended content can be pushed to the user through step S103, so that the user's terminal device can display the weather information and target recommended content. Specifically, the display method of the weather information and target recommended content can be set according to actual needs. In the embodiment of this specification, it can be achieved in the following way: construct a target push interface, the target push interface includes a first area and a second area; fill the weather information in the first area, and fill the brief information of the target recommended content in the second area; push the target push interface to the user, so that the user's terminal device displays the target push interface.

[0112] In the specific implementation process, the positional relationship between the first area and the second area can be set according to actual needs, and is not limited here. In the embodiment of this specification, the second area is located below the first area as an example for explanation. Figure 2 The figure is a schematic diagram of the target push interface provided in the embodiment of this specification. Figure 2 In the example, the weather information in the first area includes weather information such as temperature and wind speed, as well as information such as life index, and the second area includes brief information of the target recommended content. The brief information of the target recommended content may include, but is not limited to, the title and summary of the target recommended content.

[0113] The display method of the target recommended content in the second area can be set according to actual needs, for example, it can be presented in the form of hot searches, in the form of titles plus content-related pictures, etc. Figure 2 As shown, the content is presented in the form of hot searches. In addition, due to the limited display area of ​​the second area, all target recommended content may not be displayed. Therefore, a jump entry for displaying more target recommended content may be provided in the second area. In a specific implementation, when a target operation for the jump entry is received, a secondary display interface is pushed to the user, wherein the secondary display interface is used to display other recommended content, and the other recommended content is different from the target recommended content displayed in the second area.

[0114] In the specific implementation process, Figure 2 As shown in the figure, a total of 7 target recommended contents are displayed in the second display area. At the same time, the second display area also displays a jump entrance of "More Hot Searches". Users can view other target recommended contents by clicking "More Hot Searches". It should be noted that the 7 target recommended contents on the target push interface can be the recommended contents that best meet the user's search habits. Clicking one of the titles can jump to the details page of the target recommended content corresponding to the title. Figure 2If the first recommended content in it is "Yellow Gale Warning in Haidian, Beijing", then it jumps to the details display interface of this recommended content. Additionally, after clicking "More Hot Searches", a secondary display interface can be entered to display a list of other target recommended contents. The display method of the target recommended contents in the secondary interface can be the same as or different from that in the target push interface, which is not limited here.

[0115] In the embodiments of this specification, tags can also be added to each target recommended content, such as tags representing local content, tags representing popular topics, tags representing recommended readings, etc. At the same time, a corresponding display icon is set for each tag. For example, the display icon for the local content tag is "Local", the display icon for the popular topic tag is "Hot", and the display icon for the recommended reading tag is "Rec". When displaying, the display icon of the tag can be displayed behind each target recommended content.

[0116] In addition, when presenting the target recommended contents, they can also be classified and displayed according to the category tags to which the target recommended contents belong, and each category tag can correspond to its own display tab. For example, the category tags to which the target recommended contents belong can include morning life news, fashion and entertainment, sports, etc., then the display area of the target recommended contents can include multiple tabs, and by clicking on the tabs, switching between recommended contents of different categories can be performed.

[0117] In the embodiments of this specification, personalized content recommendation can be carried out during the high-frequency search period corresponding to the weather search demand, that is, during the high-frequency period of weather search, the target push interface is pushed to the user to simultaneously display weather information and target recommended contents. And during other periods except the above high-frequency period, or during the non-high-frequency search period corresponding to the weather search demand, if a weather search request is received, in order to save computing resources, only the weather information corresponding to the target location can be obtained, and the weather display interface is pushed to the user, where the weather display interface is only used to display weather information. It should be noted that the non-high-frequency search period corresponding to the weather search demand refers to the time period in a day when the user traffic for weather search is less than or equal to the preset traffic.

[0118] It should be understood that for applications such as search engines that can obtain a large amount of recommended contents, in order to quickly determine the contents that meet the user's preferences from the large amount of recommended contents, the large amount of recommended contents can be initially filtered first, filtering out the contents that the user is not interested in and leaving the contents that the user is interested in, so as to further perform fine screening on the contents that the user is interested in.

[0119] Therefore, in the embodiments of this specification, the obtaining of the initial recommended content set may include: screening out the initial recommended content set in a preset recommended content library based on the attribute information of the user.

[0120] Specifically, the user's attribute information may include the user's permanent location, the user's search habits, the user's portrait tags and other information. The preset recommended content library may be all data that can be used as recommended content. The preset recommended content library may include historical releases and data from various fields released on the day of the weather search, such as news, finance, sports, music, movies, beauty, health and other fields, articles, videos, audio, pictures, etc. The preset recommended content library is preliminarily filtered to select the initial recommended content set as the content to be recommended to the user. When selecting the initial recommended content set, it can be selected according to the user's personalized characteristics and user needs. Below, several methods for selecting the initial recommended content set are specifically described.

[0121] The first method: determine the user's permanent location; if the target location is different from the permanent location, filter out the recommended content corresponding to the target location from the preset recommended content library as the initial recommended content set; if the target location is the same as the permanent location, filter out the recommended content corresponding to the permanent location from the preset recommended content library as the initial recommended content set.

[0122] In the specific implementation process, when users query the weather, one is to query the weather at the permanent location, and the other is to query the weather at a different location. When users query their permanent location, their needs are usually to meet their daily travel or life needs. Generally speaking, after users query the weather at their permanent location, their searches for other keywords are usually not much different from their usual searches. Therefore, when users query the weather information at their permanent location, they can filter out the recommended content corresponding to the permanent location, such as recommended content that meets the user's daily life needs and meets the user's preferences, as the initial recommended content set, such as recommended content for local life services and areas of interest to users.

[0123] When the user inquires about the weather in a different place, that is, when the target location is different from the permanent location, it is very likely that the user needs to go on a business trip or travel to the target location. At this time, in addition to meeting travel needs, the user's query on the weather of the target location needs to meet business trip or travel needs to a large extent. Therefore, the recommended content corresponding to the target location can be used as the initial recommended content set to be recommended to the user. Among them, the recommended content corresponding to the target location may include: attraction recommendations, food recommendations, travel guides, and other content at the target location. Of course, in addition to the recommended content corresponding to the target location, you can also filter out recommended content that meets the user's preferences and add it to the initial recommended content set.

[0124] In the embodiments of this specification, the determination of the permanent location can be obtained in a variety of ways, and several of them are described below by way of example:

[0125] ① Obtain through user portraits. When portraying user portraits, add the user's permanent location label and determine the user's permanent location by obtaining the user's permanent location label.

[0126] ② By counting the location information of the user's terminal device over a period of time, the location information can be obtained through GPS (Global Positioning System). For example, the location information of the user's terminal device in the past six months or one year is obtained, and the area with the highest probability of location information appearing is counted as the user's permanent location.

[0127] ③ The registration place of the user’s mobile phone number shall be regarded as the user’s permanent place of residence. For example, if the registration place of the user’s mobile phone number is Tianjin, Tianjin shall be regarded as the user’s permanent place of residence.

[0128] ④ Determine the permanent location through the location of the user's points of interest. The specific implementation process is: obtain the location of the user's points of interest; based on the location of the points of interest, determine the user's permanent location.

[0129] Specifically, when a user performs a sharing operation through a terminal device, such as posting to Moments, Weibo, or a short video, the user can add the location of a point of interest (POI) at that time, such as the location of a store or a restaurant. By counting the locations of POIs over a period of time, such as the locations of POIs in the last three or ten months, the region to which each POI location belongs is determined, and the region with the most occurrences is used as the user's permanent location.

[0130] The second method: determine the user's historical high-frequency search terms, which are keywords that have been searched more than a preset number of times by the user within a preset time period before and / or after searching for weather during a historical period; and select recommended content corresponding to the historical high-frequency search terms from a preset recommended content library as an initial recommended content set.

[0131] Generally speaking, each user has his or her own search habits. For example, after searching for the weather, user A continues to search for today's hot news, financial news, and sports news, while user B is accustomed to searching for today's hot news first, then searching for the weather, and then searching for traffic conditions and food information. The search habits of users are basically relatively stable, that is, the keywords that users search for every day are likely to be similar. Therefore, the embodiments of this specification determine the user's historical high-frequency search terms based on the user's search habits, so that the historical high-frequency search terms are more matched with the user's search habits.

[0132] In the specific implementation process, the historical period and the preset duration can be set as needed, for example, the historical period is the past three months, the past year, etc., the preset duration can be one minute, three minutes, etc., and the preset duration before searching for weather and the preset duration after searching for weather can be the same or different. For example, the historical period can be the past six months, the preset duration before searching for weather is 1 minute, and the preset duration after searching for weather is 2 minutes.

[0133] It should be noted that the preset time in the embodiments of this specification can be selected to be a time period close to the weather search, so as to reflect the user's search habits before and after the weather search. If the preset time is far away from the weather search, the determined high-frequency search terms may only represent the user's preferences, and have little correlation with the search habits before and after the weather search. For example, user C usually searches for weather between 7:00 and 8:00 in the morning, and is accustomed to searching for hot news and traffic conditions before and after the weather search. However, the user is most keen on searching for food information outside of 7:00 to 8:00 in the morning. If the preset time is set to within 12 hours after the weather search, it is very likely that the determined historical high-frequency search terms are keywords of food information, but the keywords of food information obviously do not conform to the user's search habits when searching for weather. If food content is recommended during weather search, the user's search needs when searching for weather cannot be met.

[0134] In order to ensure accurate recommendations of content that meets the user's search habits during weather searches, in the embodiments of this specification, when counting historical high-frequency search terms within a historical period, the first keyword searched after the weather search can be recorded, and then the number of searches for these keywords can be counted, and the search terms with a search number greater than a preset number can be used as historical high-frequency search terms. It should be noted that the preset number can be set according to actual needs, for example, the preset number is 350 times, 500 times, etc. Alternatively, the keywords can be sorted according to the number of searches corresponding to each keyword, and one or more keywords with the largest number of searches can be selected as historical high-frequency search terms, for example, the five keywords with the largest number of searches can be used as historical high-frequency search terms.

[0135] After determining the historical high-frequency search terms, the recommended content corresponding to the historical high-frequency search terms is screened out from the preset recommended content library as the initial recommended content set. Specifically, it is possible to determine the classification label of the recommended content to which the high-frequency search terms belong, wherein the corresponding relationship between the search terms and the classification labels of the recommended content can be preset, or a classification model can be pre-trained, and the model is used to output the classification label corresponding to the search terms. Further, according to the category label to which the recommended content corresponding to the high-frequency search terms belongs, the content under the category label is screened out from the preset recommended content library as the initial recommended content set.

[0136] The third method: determine the target group label to which the user belongs; and select the recommended content corresponding to the target group label from the preset recommended content library as the initial recommended content set. The target group label may be a group category label of the target user group to which the user belongs, and may be used to characterize the search preferences of group users.

[0137] Specifically, users can be divided into group categories according to their interests and hobbies. There are many ways to divide groups into categories, for example, by age, gender, region, interests and hobbies, etc. In the specific implementation process, group categories can be divided by user portraits. For example, user portraits include age labels and gender labels, so users can be divided into the following groups according to age and gender: Group A is women aged 20 to 30, Group B is women aged 30 to 40, Group C is men aged 20 to 30, and Group D is men aged 30 to 40. For another example, user tags include user interest labels, so users can be divided into the following groups according to their interests and hobbies: Group E is a group that likes food, Group F is a group that likes sports, and Group G is a group that likes fashion.

[0138] In addition, in addition to the above-mentioned division of groups by user portrait labels, the similarity of user portraits between any two users can also be directly calculated. For example, if the similarity of user portraits between two users is greater than a threshold, the two users are divided into the same group. The size of the threshold can be set according to actual needs, such as 90%, 95%, etc., which is not limited here. Of course, users can also be divided into groups according to other methods, not limited to division by user portraits.

[0139] After dividing the user groups, you can further determine the group labels, where the group labels can be used to characterize the search preferences of the group users. The determination of group labels can be achieved in the following ways:

[0140] ① For each user group, the search keywords of the user group before and after searching for the weather can be used to take the category of one or more keywords that appear the most as the group label of the user group. For example, the search keywords of each user in the user group K before and after searching for the weather are counted, and the keywords that appear the most are "cat", "corgi", "pet dog feeding", and "cat, vaccine". In order to determine the category to which these keywords belong, a keyword classification model can be pre-trained, and the keywords can be used as the input of the model, and the corresponding category can be output as the group label of the user group. For example, "cat", "corgi", "pet dog feeding", and "cat, vaccine" are input into the keyword classification model, and the corresponding category is "cute pet", then the group label of the user group is "cute pet".

[0141] ② Determine the user profile of each user in the user group, and count the one or more tags that appear most frequently in the user profile as the tag of the user group. For example, in user group S, the tags that appear most frequently in the user profile are "sports" and "outdoors", so "sports" and "outdoors" can be used as the group tags of the user group.

[0142] Of course, group labels can also be determined in other ways, which are not listed here.

[0143] In the embodiments of this specification, after determining the target group label to which the user belongs, the recommended content corresponding to the target group label is screened out in the preset recommended content library. In the specific implementation process, the content in the preset recommended content library can be classified according to the existing group labels. For example, the group labels include eight labels, namely, "real things", "finance", "health", "fashion", "entertainment", "sports", "cute pets", and "celebrity". Then, each recommended content in the preset recommended content library can be mapped to at least one of the labels, that is, the recommended content of the preset recommended content library is divided into eight categories. In this way, after determining the target group label, for example, if the target group labels are "finance" and "health", the recommended content belonging to the two categories of "finance" and "health" in the preset recommended content library can be screened out as the initial recommended content set.

[0144] It should be noted that, in a specific implementation process, the above-mentioned several methods for screening the initial recommended content set may be combined with each other or performed separately, which is not limited here.

[0145] In the embodiments of this specification, considering that the identity of the user may be certain or may not be certain when the user searches for weather, one or more user portraits may be created for the user. Specifically, the user portrait may include: a user portrait corresponding to the user identifier, a user portrait corresponding to the user's terminal device, and a user portrait corresponding to the user identifier and the terminal device.

[0146] Specifically, before querying the weather, the user logs in to the search engine first. Then, the user ID used to characterize the user's identity can be uniquely determined based on the user's login information. At this time, the user portrait corresponding to the user ID can be selected for content recommendation. If the user has not logged in to the search engine, and the user identity cannot be determined by other means, in order to accurately make personalized recommendations, at this time, the user portrait of the terminal device currently used by the user can be selected for content recommendation. In addition, considering that the user may have logged in to other users' accounts, in order to avoid using other users' user portraits as the current user's portrait, the accuracy of the user portrait can be ensured by using a user portrait corresponding to both the user ID and the terminal device.

[0147] Based on this, the method for obtaining the target portrait in the embodiments of this specification can be: obtaining the target identifier corresponding to the user, the target identifier including at least one of the user identifier and the device identifier; based on the preset correspondence between the identifier and the user portrait, determining the target user portrait corresponding to the target identifier as the target portrait.

[0148] In the specific implementation process, obtaining the target identifier corresponding to the user may include three situations: obtaining the user identifier alone; obtaining the device identifier alone; and obtaining the user identifier and the device identifier at the same time. Correspondingly, the preset correspondence between the identifier and the user portrait also includes three situations: the preset correspondence between the user identifier and the user portrait; the preset correspondence between the device identifier and the user portrait; and the preset correspondence between the user identifier, the device identifier, and the user portrait. The following describes these three situations in detail.

[0149] The first case: the target identifier is the user identifier, and the preset corresponding relationship is the preset corresponding relationship between the user identifier and the user portrait. In the specific implementation process, the target portrait of the user can be determined in the following manner: obtaining the target user identifier of the user; based on the preset corresponding relationship between the user identifier and the user portrait, determining the target user portrait corresponding to the target user identifier as the target portrait.

[0150] It should be noted that the user identity in this case can be determined. The user identity can be determined by detecting the user's login status, which can be the login status of the current application used by the user to search for weather, or the login status of other applications. For example, taking the current application as a search engine as an example, the user searches for weather through a search engine. At this time, it can be detected whether the user is logged in to the search engine. If in a logged-in state, the target user ID can be determined based on the user's login information. The target user ID can be the user's login account, or it can be a user ID individually assigned to each user.

[0151] If it is detected that the user has not logged into the current application, such as the search engine, it can be further detected whether the user has other login information associated with the current application account. For example, the user's terminal device can record and manage the login information of each application running on the terminal device. That is, the login information of each application is associated with the user, and when it is detected that the user has logged into one of the applications, the login information of other applications can be found accordingly. For example, the user has not logged into the search engine, but the user has logged into the network disk. Then, based on the login information of the network disk and the above-mentioned associated login information, the user's search engine login information can be obtained, and then the user ID can be determined based on the search engine login information. Alternatively, the login information of the same user in each application can also be matched with the user ID of the user, so that as long as it is detected that the user has logged into one of the applications, the user ID of the user can be obtained.

[0152] The server may store a preset correspondence between user identifiers and user portraits. The user portrait of each user identifier in the preset correspondence is pre-built. Specifically, for a user identifier, taking the user identifier as a login account of a search engine as an example, when the user is logged in, the user's data in multiple dimensions is collected, such as user attribute information, searched keywords, viewed content, browsing time of each content, etc. A user tag of the user is generated based on these data, and the user tag is matched with the user identifier to obtain a user portrait of the user. Since the user portrait in this case is obtained when the user identity is determined, the user portrait corresponding to the user identifier can accurately portray the user's preferences.

[0153] After determining the user's target user ID, the target user portrait of the user can be determined by querying the preset correspondence between the user ID and the user portrait. Personalized recommendations based on the target user portrait can be more in line with the user's search habits and preferences.

[0154] The second case: the target identifier is the device identifier, and the preset correspondence is the preset correspondence between the device identifier and the user portrait. Specifically, considering that some users tend to use the functions provided by various applications without logging in, in order to provide accurate personalized recommendations to these users, in the embodiments of this specification, when the user identity is not determined, the user's target portrait can be determined in the following manner: obtaining the target device identifier of the terminal device that receives the weather search request; based on the preset correspondence between the device identifier and the device user portrait, determining the target device user portrait corresponding to the target device identifier as the target portrait.

[0155] Specifically, a device identifier is an identifier used to uniquely characterize a device. A user's device, such as a smartphone, is usually used by the user himself. Therefore, the usage of the device can also reflect the preferences and habits of the owner of the device to a certain extent. In the embodiments of this specification, by collecting data of multiple dimensions during the use of the device, such as searched keywords, viewed content, browsing time, frequency of use of each application, usage time of each application, etc., a portrait corresponding to the device can be generated based on these data, that is, a device user portrait corresponding to the device identifier.

[0156] When a user performs a weather search, if the identity of the user cannot be determined, the target device identifier of the device receiving the user's weather search input is determined, and the target device user portrait is determined by querying the preset correspondence between the device identifier and the device user portrait, and personalized content recommendations are performed based on the target device user portrait.

[0157] The third case: the target identifier is the user identifier and the device identifier, and the preset corresponding relationship is the preset corresponding relationship between the user identifier, the device identifier and the user portrait.

[0158] Specifically, when a user is logged in, he or she may be logged into another user's account. For example, user A logs into user B's search engine account on his or her mobile phone and searches for weather. If the user profile is determined only by the user ID, since the logged in account is user B's account, the obtained user profile is that of user B, while the actual user of the mobile phone is user A. It is obviously unreasonable to make personalized recommendations for user A based on user B's user profile.

[0159] Therefore, in the embodiment of the present specification, when the user identification can be obtained, it is first determined whether the user identification is the identification of the current user who performs the weather search. Specifically, the target user identification currently in the login state is obtained, and the first portrait corresponding to the target user identification is determined according to the method described in the first case above; at the same time, the target device identification of the terminal device that receives the weather search request is obtained, and the second portrait corresponding to the target device identification is determined according to the method described in the second case above; the similarity between the first portrait and the second portrait is calculated. Generally speaking, if the logged-in user is the owner of the terminal device, the similarity between the first portrait and the second portrait is very high. If the similarity between the first portrait and the second portrait is greater than the preset similarity, the user identification is considered to be the identification of the current user who performs the weather search. It should be noted that the preset similarity can be selected according to actual needs, such as 90%, 95%, etc., which is not limited here. At this time, the preset correspondence between the user identification, the device identification and the user portrait can be the preset correspondence between the user identification and the user portrait, or the preset correspondence between the device identification and the user portrait, or the preset correspondence between the pre-built corresponding to both the user identification and the device identification.

[0160] Specifically, the pre-built preset correspondence corresponding to both the user ID and the device ID can be generated when the user is logged in, for example, when the user is logged in to the search engine, by collecting data in multiple dimensions during the user's use of the search engine and collecting data in multiple dimensions during the use of the device. The portrait corresponding to both the user ID and the device ID can accurately portray the user's usage habits.

[0161] On the other hand, if the similarity between the first portrait and the second portrait is less than the preset similarity, it indicates that the user identification may not correspond to the current user performing the search operation. At this time, a prompt box may pop up to ask the user whether the currently logged-in account is his own account. If so, the preset correspondence between the user identification, device identification and user portrait can be the preset correspondence between the user identification and the user portrait, or it can be the above-mentioned pre-built preset correspondence corresponding to both the user identification and the device identification. If not, the preset correspondence between the user identification, device identification and user portrait can be the preset correspondence between the device identification and the user portrait. Alternatively, if the similarity between the first portrait and the second portrait is less than the preset similarity, the preset correspondence between the user identification, device identification and user portrait can be set by default to one of the preset correspondence between the user identification and the user portrait and the preset correspondence between the device identification and the user portrait.

[0162] To sum up, the method provided in the embodiments of the present specification can make personalized content recommendations to users based on user portraits based on users' weather search requests. On the one hand, since user portraits can reflect users' search habits and preferences, the target recommended content determined based on target portraits will be more in line with users' search preferences and user needs. On the other hand, users do not need to repeatedly search for keywords, which effectively reduces users' search costs.

[0163] Based on the same inventive concept, the embodiment of this specification also provides a recommendation device based on weather search, such as Figure 3 As shown, the device comprises:

[0164] The weather information determination module 501 is used to obtain weather information corresponding to the target location when receiving a weather search request from a user for a target location within a high-frequency search period corresponding to the weather search demand, wherein the high-frequency period corresponding to the weather search demand refers to a period of time in a day when the user flow for weather search is greater than a preset flow;

[0165] The recommended content determination module 502 is used to obtain the target profile of the user and determine the target recommended content based on the target profile;

[0166] The push module 503 is used to push the weather information corresponding to the target location and the target recommended content to the user.

[0167] In an optional embodiment, the recommended content determination module 502 is used to: obtain an initial recommended content set, determine the attribute information of each recommended content in the initial recommended content set; score the relevance of each recommended content to the target portrait based on the target portrait and the attribute information of each recommended content; and select a preset number of recommended content with larger scores as the target recommended content.

[0168] In an optional implementation, the attribute information includes at least one of the following attributes: a category label corresponding to the recommended content, a historical click rate of the recommended content, a publishing source of the recommended content, a publishing time of the recommended content, and a content value score of the recommended content.

[0169] In an optional embodiment, the push module 503 is used to: construct a target push interface, the target push interface including a first area and a second area; fill the weather information in the first area, and fill the brief information of the target recommended content in the second area; push the target push interface to the user so that the user's terminal device displays the target push interface.

[0170] In an optional implementation manner, a jump entrance is further provided in the second area, and the device further comprises:

[0171] A secondary interface push module is used to push a secondary display interface to the user when receiving a target operation for the jump entry; wherein the secondary display interface is used to display other recommended content, and the other recommended content is different from the target recommended content displayed in the second area.

[0172] In an optional embodiment, the device further comprises:

[0173] an acquisition module, configured to acquire weather information corresponding to the target location upon receiving the weather search request within a non-high-frequency search period corresponding to the weather search demand, wherein the non-high-frequency period corresponding to the weather search demand refers to a period of time within a day when the user traffic for weather search is less than or equal to the preset traffic;

[0174] A weather push module is used to push a weather display interface to the user, wherein the weather display interface is used to display the weather information.

[0175] In an optional implementation, the recommended content processing module 502 is configured to: based on the attribute information of the user, filter out the initial recommended content set from a preset recommended content library.

[0176] In an optional embodiment, the recommended content processing module 502 is used to: determine the user's permanent location; if the target location is different from the permanent location, filter out the recommended content corresponding to the target location in the preset recommended content library as the initial recommended content set; if the target location is the same as the permanent location, filter out the recommended content corresponding to the permanent location in the preset recommended content library as the initial recommended content set.

[0177] In an optional embodiment, the recommended content processing module 502 is used to: determine the user's historical high-frequency search terms, the historical high-frequency search terms are keywords that the user searched for with a frequency greater than a preset frequency within a preset time period before and / or after searching for weather within a historical period; and filter out recommended content corresponding to the historical high-frequency search terms in the preset recommended content library as the initial recommended content set.

[0178] In an optional implementation, the recommended content processing module 502 is used to: determine a target group label of a target user group to which the user belongs; and filter out recommended content corresponding to the target group label in the preset recommended content library as the initial recommended content set.

[0179] In an optional embodiment, the recommended content processing module 502 is used to: obtain a target identifier corresponding to the user, the target identifier including at least one of a user identifier and a device identifier; based on a preset correspondence between the identifier and the user portrait, determine a target user portrait corresponding to the target identifier as the target portrait.

[0180] Regarding the device in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0181] Based on the same inventive concept, an embodiment of the present invention provides an electronic device 800, Figure 4 8 is a block diagram of an electronic device 800 according to an exemplary embodiment. For example, the device 800 may be a mobile phone, a computer, a digital broadcast terminal, a messaging device, a game console, a tablet device, a medical device, a fitness device, a personal digital assistant, etc.

[0182] Reference Figure 4 , device 800 may include one or more of the following components: a processing component 802 , a memory 804 , a power component 806 , a multimedia component 808 , an audio component 810 , an input / output (I / O) interface 812 , a sensor component 814 , and a communication component 816 .

[0183] The processing component 802 generally controls the overall operation of the device 800, such as operations associated with display, phone calls, data communications, camera operations, and recording operations. The processing component 802 may include one or more processors 820 to execute instructions to complete all or part of the steps of the above-mentioned method. In addition, the processing component 802 may include one or more modules to facilitate the interaction between the processing component 802 and other components. For example, the processing component 802 may include a multimedia module to facilitate the interaction between the multimedia component 808 and the processing component 802.

[0184] The memory 804 is configured to store various types of data to support operations on the device 800. Examples of such data include instructions for any application or method operating on the device 800, contact data, phone book data, messages, pictures, videos, etc. The memory 804 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0185] The power component 806 provides power to the various components of the device 800. The power component 806 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power for the device 800.

[0186] The multimedia component 808 includes a screen that provides an output interface between the device 800 and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen may be implemented as a touch screen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touch, slide, and gestures on the touch panel. The touch sensor may not only sense the boundaries of the touch or slide action, but also detect the duration and pressure associated with the touch or slide operation. In some embodiments, the multimedia component 808 includes a front camera and / or a rear camera. When the device 800 is in an operating mode, such as a shooting mode or a video mode, the front camera and / or the rear camera may receive external multimedia data. Each front camera and the rear camera may be a fixed optical lens system or have a focal length and optical zoom capability.

[0187] The audio component 810 is configured to output and / or input audio signals. For example, the audio component 810 includes a microphone (MIC), and when the device 800 is in an operating mode, such as a call mode, a recording mode, and a speech recognition mode, the microphone is configured to receive an external audio signal. The received audio signal can be further stored in the memory 804 or sent via the communication component 816. In some embodiments, the audio component 810 also includes a speaker for outputting audio signals.

[0188] I / O interface 812 provides an interface between processing component 802 and peripheral interface modules, such as keyboards, click wheels, buttons, etc. These buttons may include but are not limited to: home button, volume button, start button, and lock button.

[0189] The sensor assembly 814 includes one or more sensors for providing various aspects of status assessment for the device 800. For example, the sensor assembly 814 can detect the open / closed state of the device 800, the relative positioning of components, such as the display and keypad of the device 800, and the sensor assembly 814 can also detect the position change of the device 800 or a component of the device 800, the presence or absence of user contact with the device 800, the orientation or acceleration / deceleration of the device 800, and the temperature change of the device 800. The sensor assembly 814 may include a proximity sensor configured to detect the presence of nearby objects without any physical contact. The sensor assembly 814 may also include a light sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, the sensor assembly 814 may also include an accelerometer, a gyroscope sensor, a magnetic sensor, a pressure sensor, or a temperature sensor.

[0190] The communication component 816 is configured to facilitate wired or wireless communication between the device 800 and other devices. The device 800 can access a wireless network based on a communication standard, such as WiFi, 2G or 3G, or a combination thereof. In an exemplary embodiment, the communication component 816 receives a broadcast signal or broadcast-related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component 816 also includes a near field communication (NFC) module to facilitate short-range communication. For example, the NFC module can be implemented based on radio frequency identification (RFID) technology, infrared data association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology and other technologies.

[0191] In an exemplary embodiment, the device 800 may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the above methods.

[0192] In an exemplary embodiment, a non-transitory computer-readable storage medium including instructions is also provided, such as a memory 804 including instructions, and the instructions can be executed by the processor 820 of the device 800 to perform the above method. For example, the non-transitory computer-readable storage medium can be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, etc.

[0193] Figure 51 is a schematic diagram of the structure of the server in some embodiments of the present invention. The server 1900 may have relatively large differences due to different configurations or performances, and may include one or more central processing units (CPU) 1922 (for example, one or more processors) and memory 1932, and one or more storage media 1930 (for example, one or more mass storage devices) storing application programs 1942 or data 1944. Among them, the memory 1932 and the storage medium 1930 can be short-term storage or permanent storage. The program stored in the storage medium 1930 may include one or more modules (not shown in the figure), and each module may include a series of instruction operations in the server. Furthermore, the central processing unit 1922 can be configured to communicate with the storage medium 1930 and execute a series of instruction operations in the storage medium 1930 on the server 1900.

[0194] The server 1900 may also include one or more power supplies 1926, one or more wired or wireless network interfaces 1950, one or more input and output interfaces 1958, one or more keyboards 1956, and / or, one or more operating systems 1941, such as Windows ServerTM, Mac OS XTM, UnixTM, LinuxTM, FreeBSDTM, etc.

[0195] A non-temporary computer-readable storage medium, when the instructions in the storage medium are executed by a processor of a device (server or terminal), enables the device to execute the data processing method described in the aforementioned embodiments.

[0196] A non-transitory computer-readable storage medium, when the instructions in the storage medium are executed by a processor of a device (server or terminal), enables the device to perform a recommendation method based on weather search, the method comprising:

[0197] When a user's weather search request for a target location is received within a high-frequency search period corresponding to a weather search demand, weather information corresponding to the target location is obtained, wherein the high-frequency period corresponding to the weather search demand refers to a period of time in a day when the user flow for weather search is greater than a preset flow;

[0198] Obtaining a target profile of the user, and determining target recommended content based on the target profile;

[0199] The weather information corresponding to the target location and the target recommended content are pushed to the user.

[0200] One or more technical solutions provided by the embodiments of the present invention achieve at least the following technical effects or advantages:

[0201] After receiving the user's weather search request during the high-frequency search period corresponding to the weather search needs, personalized content recommendations can be made to the user based on the user portrait. On the one hand, since the user portrait can reflect the user's search habits and preferences, the target recommended content determined based on the target portrait will be more in line with the user's search preferences and user needs. On the other hand, there is no need for the user to repeatedly search for keywords, which effectively reduces the user's search cost.

[0202] Those skilled in the art will readily appreciate other embodiments of the invention after considering the specification and practicing the invention disclosed herein. The present invention is intended to cover any variations, uses or adaptations of the present invention that follow the general principles of the present invention and include common knowledge or customary techniques in the art that are not disclosed in this disclosure. The description and examples are to be considered exemplary only, and the true scope and spirit of the present invention are indicated by the following claims.

[0203] It should be understood that the present invention is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from the scope thereof. The scope of the present invention is limited only by the appended claims. The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modification, equivalent substitution, improvement, etc. made within the spirit and principle of the present invention shall be included in the scope of protection of the present invention.

Claims

1. A recommendation method based on weather search, characterized in that: include: When a user's weather search request for a target location is received within a high-frequency search period corresponding to a weather search demand, weather information corresponding to the target location is obtained, wherein the high-frequency period corresponding to the weather search demand refers to a period of time in a day when the user flow for weather search is greater than a preset flow; When the weather search request is received during the high-frequency search period, the target profile of the user is obtained, and target recommended content is determined based on the target profile, including: obtaining an initial recommended content set, and determining attribute information of each recommended content in the initial recommended content set; scoring the relevance of each recommended content to the target profile based on the target profile and the attribute information of each recommended content; and selecting a preset number of recommended content with larger scores as the target recommended content; Pushing weather information corresponding to the target location and the target recommended content to the user; The obtaining of the initial recommended content set includes: Based on the attribute information of the user, the initial recommended content set is screened out from a preset recommended content library, wherein the historical high-frequency search terms of the user are determined, and the historical high-frequency search terms are keywords whose search frequency is greater than a preset frequency within a preset time period before and / or after the weather search by the user in a historical period; and recommended content corresponding to the historical high-frequency search terms is screened out from the preset recommended content library as the initial recommended content set; The pushing the weather information corresponding to the target location and the target recommended content to the user includes: During the high-frequency search period, a target push interface is constructed, the target push interface comprising a first area and a second area; the weather information is filled in the first area, and the brief information of the target recommended content is filled in the second area; the target push interface is pushed to the user so that the user's terminal device displays the target push interface, wherein the target recommended content is presented in the second area in the form of a hot search, and a label is added to each target recommended content; During a non-high-frequency search period, a weather display interface for displaying the weather information is pushed to the user.

2. The method according to claim 1, characterized in that The step of selecting the initial recommended content set from a preset recommended content library based on the attribute information of the user includes: determining the user's usual place of residence; If the target location is different from the permanent location, selecting recommended content corresponding to the target location from the preset recommended content library as the initial recommended content set; If the target location is the same as the permanent location, the recommended content corresponding to the permanent location is selected from the preset recommended content library as the initial recommended content set.

3. The method according to claim 1, characterized in that: The step of selecting the initial recommended content set from a preset recommended content library based on the attribute information of the user includes: Determine a target group label of a target user group to which the user belongs; The recommended content corresponding to the target group label is screened out from the preset recommended content library as the initial recommended content set.

4. A weather search-based recommendation device, characterized in that: include: A weather information processing module, configured to obtain weather information corresponding to a target location when receiving a weather search request from a user within a high-frequency search period corresponding to a weather search demand, wherein the high-frequency period corresponding to the weather search demand refers to a period of time in a day when the user flow for weather search is greater than a preset flow; The recommended content processing module is used to obtain the target portrait of the user when receiving the weather search request in the high-frequency search period, and determine the target recommended content based on the target portrait, including: obtaining an initial recommended content set, and determining the attribute information of each recommended content in the initial recommended content set; scoring the relevance of each recommended content to the target portrait based on the target portrait and the attribute information of each recommended content; and selecting a preset number of recommended contents with larger scores as the target recommended contents; A push module, used to push the weather information corresponding to the target location and the target recommended content to the user; The recommended content processing module is further configured to filter out the initial recommended content set from a preset recommended content library based on the attribute information of the user, wherein the historical high-frequency search terms of the user are determined, and the historical high-frequency search terms are keywords whose search frequency is greater than a preset frequency within a preset time period before and / or after the user searches for weather in a historical period; and the recommended content corresponding to the historical high-frequency search terms is filtered out from the preset recommended content library as the initial recommended content set; The push module is also used to: during the high-frequency search period, construct a target push interface, the target push interface including a first area and a second area; fill the weather information in the first area, and fill the brief information of the target recommended content in the second area; push the target push interface to the user so that the user's terminal device displays the target push interface, wherein the target recommended content is presented in the second area in the form of a hot search, and a label is added to each target recommended content; during the non-high-frequency search period, push a weather display interface for displaying the weather information to the user.

5. The device according to claim 4, characterized in that The recommended content processing module is used to: determining the user's usual place of residence; If the target location is different from the permanent location, selecting recommended content corresponding to the target location from the preset recommended content library as the initial recommended content set; If the target location is the same as the permanent location, the recommended content corresponding to the permanent location is selected from the preset recommended content library as the initial recommended content set.

6. The device according to claim 4, characterized in that The recommended content processing module is used to: Determine a target group label of a target user group to which the user belongs; The recommended content corresponding to the target group label is screened out from the preset recommended content library as the initial recommended content set.

7. An electronic device, characterized in that: It comprises a memory and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by one or more processors to execute the operation instructions corresponding to the method according to any one of claims 1 to 3 contained in the one or more programs.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps corresponding to the method according to any one of claims 1 to 3 are implemented.

Images