CN111414509B

CN111414509B - Method and device for providing picture material for applet

Info

Publication number: CN111414509B
Application number: CN202010201754.9A
Authority: CN
Inventors: 江少华
Original assignee: Alipay Hangzhou Information Technology Co Ltd
Current assignee: Alipay Hangzhou Information Technology Co Ltd
Priority date: 2020-03-20
Filing date: 2020-03-20
Publication date: 2023-06-02
Anticipated expiration: 2040-03-20
Also published as: CN111414509A

Abstract

The embodiment of the specification discloses a method and a device for providing picture materials for applets. The method comprises the following steps: acquiring information of an applet, and mining keywords related to the applet from the information of the applet; vectorizing the keywords to obtain vectors of the keywords; calculating the similarity of the vector of the keyword and the vectors of the plurality of pictures; and selecting a picture serving as the image material of the applet from a plurality of pictures according to the similarity.

Description

Method and device for providing picture material for applet

Technical Field

The present specification relates to software technology, and more particularly, to a method of providing picture material for an applet and an apparatus for providing picture material for an applet.

Background

With the development of software technology and network technology, more and more applets are presented. An applet is a program that can be directly loaded for use (i.e., search and use) without downloading, installing, and searching. The applet can be carried on a native application or a web application, wherein the native application is an application program directly running on a current operating system, the web application refers to a web application needing to run in a browser, and the applet can realize own functions by depending on the native application and the web application.

Typically, a native application or web application will provide an applet management platform for a number of different applets to be provided to a user for use, which may belong to different operators, and in some cases, the platform may need to manage the manner in which these applets are presented.

Disclosure of Invention

Embodiments disclosed herein provide methods and apparatus for providing picture material for applets.

According to a first aspect of the present disclosure, there is provided a method of providing picture material for an applet, comprising the steps of:

acquiring information of an applet, wherein the information of the applet comprises names, categories and description information of the applet;

mining keywords about the applet from the applet information;

vectorizing the keywords to obtain vectors of the keywords;

calculating the similarity of the vector of the keyword and the vector of the tag information of the plurality of pictures;

and selecting a picture serving as the image material of the applet from a plurality of pictures according to the similarity.

Optionally, selecting a picture as the image material of the applet from a plurality of the pictures according to the similarity, including:

And configuring the picture with the highest similarity as an icon or a background picture of the applet.

Optionally, the information of the applet further includes information of a main page presentation of the applet.

Optionally, vectorizing the keywords to obtain vectors of the keywords, including:

and carrying out embedded vector calculation on the keywords through a BERT algorithm to obtain the vectors of the keywords.

Optionally, the vector of label information of the picture is obtained by:

acquiring label information of the picture;

and carrying out embedded vector calculation on the label information of the picture through a BERT algorithm to obtain a vector of the label information of the picture.

Optionally, calculating the similarity between the vector of the keyword and the vector of the tag information of the plurality of pictures includes:

calculating cosine similarity of the vector of the keyword and the vector of the tag information of the plurality of pictures; or alternatively, the process may be performed,

and calculating the Euclidean distance between the vector of the keyword and the vector of the tag information of the plurality of pictures.

According to a second aspect of the present disclosure, there is provided a method of providing picture material for an applet, comprising the steps of:

acquiring information of a plurality of applets, wherein the information of the applets comprises names, categories and description information of the applets;

Mining common keywords about the plurality of applets from the information of the plurality of applets;

vectorizing the common keywords to obtain vectors of the common keywords;

calculating the similarity of the vector of the common key word and the vector of the tag information of the plurality of pictures;

and selecting the picture serving as the image material of the aggregation page of the applets from the pictures according to the similarity.

Optionally, selecting, according to the similarity, a picture of the image material of the aggregate page of the plurality of applets from the plurality of pictures, including:

and configuring the picture with the highest similarity as an icon or a background picture of an aggregation page of the plurality of applets.

Optionally, vectorizing the common keywords to obtain vectors of the common keywords, including:

and performing embedded vector calculation on the common keywords through a BERT algorithm to obtain vectors of the common keywords.

Optionally, the vector of label information of the picture is obtained by:

Acquiring label information of the picture;

Optionally, calculating the similarity between the vector of the common keyword and the vector of the tag information of the plurality of pictures includes:

calculating cosine similarity of the vector of the common key word and the vector of the tag information of the plurality of pictures; or alternatively, the process may be performed,

and calculating Euclidean distance between the vector of the common keyword and the vector of the tag information of the plurality of pictures.

According to a third aspect disclosed in the present specification, there is provided an apparatus for providing picture material for an applet, comprising:

the keyword extraction module is used for acquiring information of the applet, wherein the information of the applet comprises names, categories and description information of the applet; mining keywords about the applet from the applet information;

the vectorization processing module is used for vectorizing the keywords to obtain vectors of the keywords;

the similarity calculation module is used for calculating the similarity of the vector of the keyword and the vector of the tag information of the plurality of pictures;

and the selecting module is used for selecting the picture serving as the image material of the applet from the plurality of pictures according to the similarity.

According to a fourth aspect disclosed in the present specification, there is provided an apparatus for providing picture material for an applet, comprising:

the keyword extraction module is used for acquiring information of a plurality of applets, wherein the information of the applets comprises names, categories and description information of the applets; mining common keywords about the plurality of applets from the information of the plurality of applets;

the vectorization processing module is used for vectorizing the common keywords to obtain vectors of the common keywords;

the similarity calculation module is used for calculating the similarity of the vector of the common key word and the vector of the tag information of the plurality of pictures;

and the selecting module is used for selecting pictures serving as image materials of the aggregation pages of the applets from the pictures according to the similarity.

According to a fifth aspect of the present disclosure there is provided an apparatus for providing picture material for an applet, comprising a processor and a memory, the memory storing a computer program which when executed by the processor implements the method disclosed in the first or second aspect of the present disclosure.

Features of the embodiments of the present specification and their advantages will become apparent from the following detailed description of exemplary embodiments of the present specification with reference to the accompanying drawings.

Drawings

The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments of the specification and, together with the description, serve to explain the principles of the embodiments of the specification.

FIG. 1 is a schematic diagram of an applet management system provided in an embodiment of the present disclosure;

FIG. 2 is a flow chart of a method for providing picture material for an applet provided in an embodiment of the present disclosure;

FIG. 3 is a flow chart of a method for providing picture material for an applet provided in an embodiment of the present disclosure;

FIG. 4 is a block diagram of an apparatus for providing picture material for applets, provided by an embodiment of the present disclosure;

FIG. 5 is a block diagram of an apparatus for providing picture material for applets, provided by an embodiment of the present disclosure;

FIG. 6 is a block diagram of an apparatus for providing picture material for applets, provided by an embodiment of the present disclosure;

fig. 7 (a) -7 (e) are schematic diagrams of applet related pages provided in the embodiments of the present specification.

Detailed Description

Various exemplary embodiments of the present specification will now be described in detail with reference to the accompanying drawings.

The following description of at least one exemplary embodiment is merely exemplary in nature and is in no way intended to limit the disclosure, its application, or uses.

It should be noted that: like reference numerals and letters denote like items in the following figures, and thus once an item is defined in one figure, no further discussion thereof is necessary in subsequent figures.

< applet management System >

Fig. 1 is a block diagram of an applet management system provided in an embodiment of the present description. As shown in fig. 1, the applet management system includes an applet providing applet management platform 101 and applets provided by a plurality of applet providers 103. The applet management platform 101 and the plurality of applet providers 103 may be communicatively coupled via a network 102.

The applet management platform 101 may be a server for managing applets, and may have various functions such as auditing, recording, online, offline, and monitoring of applets. The configuration of the server may include, but is not limited to: a processor 1011, a memory 1012, an interface device 1013, a communication device 1014, an input device 1015, and an output device 1016. The processor 1011 may include, but is not limited to, a central processing unit CPU, a microprocessor MCU, and the like. The memory 1012 may include, but is not limited to, ROM (read Only memory), RAM (random Access memory), nonvolatile memory such as a hard disk, and the like. Interface device 1013 may include, but is not limited to, a USB interface, a serial interface, a parallel interface, and the like. The communication means 1014 can be, for example, wired or wireless communication, and specifically can include WiFi communication, bluetooth communication, 2G/3G/4G/5G communication, and the like. Input devices 1015 may include, but are not limited to, a keyboard, mouse, and the like. Output devices 1016 may include, but are not limited to, a display screen or the like. The configuration of the server may also include only a part of the devices described above.

The applet provider 103 may also be a server of the applet operator, the configuration of which may include, but is not limited to: a processor 1031, a memory 1032, an interface device 1033, a communication device 1034, an input device 1035, and an output device 1036. The processor 1031 may include, but is not limited to, a central processing unit CPU, a microprocessor MCU, and the like. The memory 1032 may include, but is not limited to, ROM (read only memory), RAM (random access memory), nonvolatile memory such as a hard disk, and the like. The interface device 1033 may include, but is not limited to, a USB interface, a serial interface, a parallel interface, and the like. The communication device 1034 can be capable of wired or wireless communication, and specifically can include WiFi communication, bluetooth communication, 2G/3G/4G/5G communication, and the like, for example. The input device 1035 may include, but is not limited to, a keyboard, a mouse, and the like. The output devices 1036 may include, but are not limited to, a display screen or the like. The configuration of the server may also include only a part of the devices described above.

In one embodiment applied to this specification, applet management platform 101 is responsible for managing applets to provide a plurality of different applets to a user. The plurality of applets may have different categories, such as a game category, a financial category, a utility payment service category, a shopping category, a ticketing category, and the like. Each class of applet may include a plurality of specific applets below, e.g., a game class applet may include a plurality of different game applets. The applet for the utility payment service class may include an applet for water payment, an applet for electricity payment, an applet for gas payment, and the like.

The applet management system shown in fig. 1 is merely illustrative and is in no way meant to limit any of the embodiments of the present description, applications or uses thereof. Those skilled in the art will appreciate that while the foregoing describes a number of devices for an applet management platform and applet provider, embodiments of the present disclosure may refer to only a portion of the devices therein. Those skilled in the art can devise instructions in accordance with the schemes disclosed in the embodiments of the present specification. How the instructions control the processor to operate is well known in the art and will not be described in detail here.

< method of providing Picture Material for applet >

An applet is a program which can be directly loaded and used (namely, used immediately after searching) without downloading, installing and searching, and can be installed on a native application or a web application. A developer (e.g., a third party developer) may develop an applet that is hosted on a native application or web application using an interface provided by the native application or web application. The user may use the search functionality of the native application or web application, or based on the active presentation of the native application or web application, enter a specific applet or enter an applet center.

Fig. 7 (a) -7 (c) illustrate the page conversion process from a native application to an applet. Fig. 7 (a) shows a home page of a native application, where the native application is a payment application, and a plurality of interactive controls such as "swipe", "pay money", "collect money", "credit", "bank card", "applet" are set on the home page, and when a user clicks a certain interactive control, a function corresponding to the interactive control is executed or jumps to a page corresponding to the interactive control. When the user clicks on "applet", the user jumps to the applet center page shown in fig. 7 (b). On the applet center page, there are provided a plurality of applets such as "red pack", "pay fee", "lottery", etc., wherein the "pay fee" applet is an aggregate of a plurality of pay fee applets (hereinafter referred to as "pay fee" aggregate applet). When the user clicks on a certain applet, the user jumps to the page corresponding to the applet. For example, when the user clicks on the "pay" aggregation applet, the user jumps to the pay center page shown in fig. 7 (c). The payment center page is provided with a plurality of specific payment applets of 'mobile phone fee', 'broadband fee', 'electric fee', 'water fee', 'gas fee', 'heating fee'. When a user clicks a specific payment applet, the user jumps to a page corresponding to the specific payment applet so as to realize a corresponding payment function. Through these specific payment applets, the user can realize the functions of paying mobile phone fees, broadband fees, electric fees, water fees, fuel gas fees and heating air fees respectively.

As can be seen from fig. 7 (a) -7 (c), each applet is shown to the user in the form of a rectangular frame combined with a text label, which is not attractive enough and is not easy for the user to understand the use of the applet. By using the method for providing the picture material for the applet, which is provided by the embodiment of the specification, the proper picture can be automatically and efficiently matched with the applet, so that the page is beautified, a user can intuitively understand the purpose of the applet, and the probability of using the applet by the user is improved.

< first embodiment >

Referring to fig. 2, this embodiment provides a method for producing applet picture material, comprising the steps of:

s202, acquiring information of an applet, wherein the information of the applet comprises names, categories and description information of the applet.

The name of the applet may be the name of the applet provider to the applet, such as "XX shopping net", "XX movie ticket", "Red Package", "pay fee", "lottery", the name of the applet possibly representing the use of the applet.

The category of the applet is a category checked and confirmed by the applet management platform and is related to the application. The category of the applet may be, for example, a game category, a financial category, a utility payment service category, a shopping category, a ticket category, or the like, or may be a more subdivided category, such as a fresh-keeping category, a clothing category. The lottery drawing may belong to the category of shopping, and a commodity sample or a commodity coupon is issued to a user in a lottery drawing mode so as to enlarge the influence of the commodity and attract the user to browse and purchase related commodities.

In a specific example, an applet may correspond to multiple categories, which may be different angles of categories. For example, an applet focused on a group movie ticket sales service may have two categories, "group purchase" and "movie ticket". For example, a shopping applet for an overseas mother and infant product may have two categories, "overseas merchandise" and "mother and infant". The "red pack" applet may have two categories, game and social. The "pay" applet may have two categories, daily life and pay.

In a specific example, an applet may correspond to multiple categories having a hierarchical relationship. For example, one category of applet that sells women's clothing includes "clothing" and "women's clothing", where "women's clothing" is a sub-category under the "clothing" category. For example, one category of small procedures for selling fruit includes "fresh" and "fruit," where "fruit" is a sub-category under the "fresh" category.

Description of the applet, which may also contain description of the use of the applet, may be set by the applet provider or may be set by the applet management platform. For example, the description of the applet may be "sales counter discount women's dress". For example, the description of an applet that pays a cell phone fee may be "XX Mobile communication company". For example, the description of the applet may be "XX-initiated public welfare donation platform" and the description information embodies the use of the applet.

The applet information may also include information presented by the applet's main page. The main page of the applet is the page that the user jumps to after clicking the applet. For example, for an applet for paying a mobile phone fee, an input box labeled "mobile phone number" and an input box labeled "recharge amount" may be provided on a main page, so that a user may input the mobile phone number and recharge amount that the user wants to recharge, where "mobile phone number" and "recharge amount" are information displayed on the main page. It can be seen that the information presented by the main page of the applet is related to the purpose of the applet.

The applet management platform stores the information of the applet or the applet provider may obtain the information of the applet.

S204, mining keywords related to the applet from the applet information.

In this step, keywords for the applet can be mined from the applet information by a keyword mining algorithm.

In a specific example, in step S204, keyword mining may be performed using a textword (text ranking) algorithm. the textword algorithm is originally an algorithm used as a ranking of web page importance, and is applied here to mine keywords of text. the textword algorithm builds a network through adjacent relations among words, then iteratively calculates the rank value of each word node, sorts the rank values, and word nodes corresponding to the rank values with the front rank value are ranked to form keywords. Specifically, in one embodiment, the process of mining the keywords of the applet from the information of the applet using the textword algorithm is as follows:

S2042, performing word segmentation and part-of-speech tagging on an original text formed by the information of the applet, filtering out stop words, and only reserving words with specified parts of speech (nouns, verbs and adjectives) as candidate words.

And S2044, constructing a candidate word graph G= (V, E), wherein V is a node set, and E is an edge set. The node set V is composed of the candidate words obtained in step S2042, and an edge between any two nodes is constructed by adopting the co-occurrence relationship, so as to obtain an edge set E.

S2046, iteratively propagating weights of all nodes in the candidate word graphs until convergence.

S2048, sorting the node weights in a reverse order, so as to obtain a plurality of most important words as candidate keywords.

S2050, marking the candidate keywords in the original text, and if adjacent phrases can be formed, taking the adjacent phrases as keywords of the applet. If adjacent phrases cannot be formed, the candidate keywords are used as keywords of the applet.

Keyword mining is performed by using a textword algorithm, the background of a corpus can be separated, and keywords of a document can be extracted by using semantic association among words in the document.

S206, vectorizing the keywords of the applet to obtain vectors of the keywords of the applet.

In one embodiment, the embedded vector (Embedding) calculation is performed on the keywords by the BERT (Bidirectional Encoder Representation from Transformers, bi-directional encoder from transformer) algorithm, resulting in vectors of the keywords of the applet. One feature of the embedded vector (Embedding) calculation is that it can be implemented to build a mapping from a high-dimensional vector to a low-dimensional vector, and the calculation amount in step S208 can be reduced by using the low-dimensional vector to characterize the keyword.

The BERT algorithm is an open-source text preprocessing method, and is a pre-training model of NLP (Natural Language Processing natural language processing). The BERT algorithm is a bi-directional encoder algorithm, and when processing a word, the information of the word before and the word after the word can be considered, so that the semantics of the context can be obtained, character-level, word-level, sentence-level and even inter-sentence relationship features can be fully described, and the computer can be helped to better understand the language like a human being. By using the BERT algorithm, subtle differences among words can be better understood, and the obtained vector can better represent the true meaning of the keywords of the applet.

S208, calculating the similarity of the vector of the keyword and the vector of the tag information of the plurality of pictures.

In step S208, the gallery has a plurality of pictures, the pictures in the gallery have labels, and the labels are text labels for the pictures. The picture provider can add a label to the picture and upload the picture to the gallery. Or, when receiving the picture, the gallery manager adds a label to the picture. A picture may have one or more labels, for example, a picture may be labeled with two labels, a "gift" and a "gift".

In the case that the picture has a tag, in step S208, tag information of the picture may be obtained, and embedded vector (Embedding) calculation is performed on the tag information of the picture by using the BERT algorithm, so as to obtain a vector of the tag information of the picture, which is used as a vector of the tag information of the picture. One feature of the embedded vector (Embedding) calculation is that it can be implemented to create a mapping from a high-dimensional vector to a low-dimensional vector, and the low-dimensional vector is used to characterize the picture, so that the calculation amount in step S208 can be reduced.

And calculating the similarity of the vector of the keyword of the applet and the vector of the tag information of the plurality of pictures, wherein the similarity of the vector of the keyword of the applet and the vector of the tag information of the picture can reflect the association degree of the keyword of the applet and the picture.

In one embodiment, the cosine similarity of the vector of the key of the applet and the vector of the tag information of the picture is calculated. Cosine similarity is the measure of similarity between two vectors in vector space by measuring the cosine value of their included angle. The range of cosine values is between [ -1,1], the closer the cosine value is to 1, the closer the direction of the vector of the keyword of the applet and the vector of the tag information of the picture is, the higher the association degree of the keyword of the applet and the picture is. The closer the cosine value is to-1, the more opposite the direction of the vector of the keyword of the applet and the vector of the tag information of the picture, the lower the degree of association of the keyword of the applet and the picture.

In one embodiment, the distance between the vector of the keyword of the applet and the vector of the tag information of the picture is calculated, and the closer the distance is, the higher the similarity of the two. For example, the euclidean distance of the vector of the keyword of the applet and the vector of the tag information of the picture is calculated. The smaller the Euclidean distance between the vector of the keyword of the applet and the vector of the tag information of the picture, the higher the similarity between the vector and the tag information of the picture, and the higher the association degree between the keyword of the applet and the picture. The larger the Euclidean distance between the vector of the keyword of the applet and the vector of the tag information of the picture, the lower the similarity between the vector and the tag information of the picture, and the lower the association degree between the keyword of the applet and the picture.

S210, selecting a picture serving as an image material of the applet from a plurality of pictures according to the similarity.

In one embodiment, the picture with the highest similarity to the keywords of the applet is used as the picture material of the applet. In one specific example, the picture with the highest similarity is configured as an icon of the applet. In a specific example, the picture with the highest similarity is configured as the background picture of the applet.

In fig. 7 (b), the icon of the "red pack" applet is in the form of a rectangular frame, and the text "red pack" is marked in the rectangular frame, which is not beautiful enough and is not beneficial for the user to understand the purpose of the applet. By using the method of the embodiment of the present specification, the keyword of the "red packet" applet includes "red packet", the tag of the "red packet" picture includes "red packet", and the similarity between the keyword of the "red packet" applet and the "red packet" picture is highest through the similarity calculation in step S208, and the "red packet" picture is selected as the image material of the "red packet" applet. In the page 7 (a), when the user clicks on the "applet", the jumped applet center page is shown in fig. 7 (d), in the page 7 (d), the icon of the "redbag" applet is a redbag picture, and after the user sees the "redbag" picture, the user intuitively understands that the "redbag" applet is used for issuing the redbag, so as to attract the user to interact with friends of the user using the "redbag" applet.

In fig. 7 (b), the icon of the "lottery" applet is in the form of a rectangular frame, and the text "lottery" is marked in the rectangular frame. By using the method of the embodiment of the specification, the keywords of the "lottery" applet comprise "gift" and "prize", the labels of the "gift" pictures comprise "gift", and the similarity between the keywords of the "lottery" applet and the "gift" pictures is highest through the similarity calculation of step S208, and the "gift" pictures are selected as the image materials of the "lottery" applet. In the page 7 (b), after clicking the "lottery" applet, the user jumps to the page shown in fig. 7 (e), and it can be seen that the "gift box" picture is configured as a background picture of the "lottery" applet, and the user clicks the interactive control "try in hand" to realize the lottery. In fig. 7 (e), matching the "lottery" applet with the "gift box" background picture can enhance the attraction to the user and enhance the probability of the user using the applet.

In one embodiment, at least two pictures are selected from high to low according to the similarity, the selected pictures are fused to obtain a fused picture, and the fused picture is used as a picture material of the applet. For example, pattern elements in the selected picture are extracted, and a fused picture is generated using these pattern elements. For example, the selected pictures are stitched together by a surf (Speeded Up Robust Features, accelerated robust feature) based image stitching algorithm to obtain a fused picture. In one specific example, the fused picture is configured as an icon of an applet. In one specific example, the fused picture is configured as a background picture of the applet.

< second embodiment >

Referring to fig. 3, this embodiment provides a method for producing applet picture material, comprising the steps of:

s302, acquiring information of a plurality of applets, wherein the information of the applets comprises names, categories and description information of the applets.

Referring to fig. 7 (b) and 7 (c), the "payment" aggregation applet includes a variety of specific payment applets. In page 7 (b), when the user clicks on the "pay" aggregation applet, the user jumps to the pay center page shown in fig. 7 (c). The payment center page is provided with a plurality of specific payment applets of 'mobile phone fee', 'broadband fee', 'electric fee', 'water fee', 'gas fee', 'heating fee'. When a user clicks a specific payment applet, the user jumps to a page corresponding to the specific payment applet so as to realize a corresponding payment function. Through these specific payment applets, the user can realize the functions of paying mobile phone fees, broadband fees, electric fees, water fees, fuel gas fees and heating air fees respectively.

S304, common keywords about the plurality of applets are mined from the information of the plurality of applets.

In this step, common keywords for the plurality of applets may be mined from the information of the plurality of applets by a keyword mining algorithm.

In a specific example, in step S304, keyword mining may be performed using a textword (text ranking) algorithm. the textword algorithm is originally an algorithm used as a ranking of web page importance, and is applied here to mine keywords of text. the textword algorithm builds a network through adjacent relations among words, then iteratively calculates the rank value of each word node, sorts the rank values, and word nodes corresponding to the rank values with the front rank value are ranked to form keywords. Specifically, in one embodiment, the process of mining common keywords from information of multiple applets using the textword algorithm is as follows:

s3042, performing word segmentation and part-of-speech tagging on an original text formed by the information of the plurality of applets, filtering out stop words, and only keeping words with specified parts of speech (nouns, verbs and adjectives) as candidate words.

S3044, constructing a candidate word graph g= (V, E), where V is a node set and E is an edge set. The node set V is composed of the candidate words obtained in step S3042, and an edge between any two nodes is constructed by adopting the co-occurrence relationship, so as to obtain an edge set E.

S3046, iteratively propagating weights of all nodes in the candidate word graphs until convergence.

S3048, sorting the node weights in a reverse order, so as to obtain a plurality of most important words as candidate keywords.

S3050, marking the candidate keywords in the original text, and if adjacent phrases can be formed, using the adjacent phrases as common keywords of a plurality of applets. If adjacent phrases cannot be formed, the candidate keywords are used as common keywords of a plurality of applets.

Referring to fig. 7 (b) and 7 (c), the "payment" aggregation applet includes a variety of specific payment applets. In page 7 (b), when the user clicks on the "pay" aggregation applet, the user jumps to the pay center page shown in fig. 7 (c). The payment center page is provided with a plurality of specific payment applets of 'mobile phone fee', 'broadband fee', 'electric fee', 'water fee', 'gas fee', 'heating fee'. The information of the "mobile phone fee" payment applet includes "mobile phone fee", "phone fee" and "payment", the information of the "broadband fee" payment applet includes "broadband", "network" and "payment", the information of the "electric fee" payment applet includes "electric fee" and "payment", the information of the "water fee" payment applet includes "water fee" and "payment", the information of the "fuel gas fee" payment applet includes "fuel gas fee" and "payment", the information of the "warm gas fee" payment applet includes "warm gas fee" and "payment", and the common keyword of these applets is found out to be "payment" through step S304.

S306, vectorizing the common keywords to obtain vectors of the common keywords.

In one embodiment, the embedded vector (Embedding) calculation is performed on the common keywords by the BERT (Bidirectional Encoder Representation from Transformers, bi-directional encoder from transformer) algorithm, resulting in vectors of the common keywords. One feature of the embedded vector (Embedding) calculation is that it can be implemented to build a mapping from a high-dimensional vector to a low-dimensional vector, and the computation amount of step S308 can be reduced by using the low-dimensional vector to characterize the common keywords.

The BERT algorithm is an open-source text preprocessing method, and is a pre-training model of NLP (Natural Language Processing natural language processing). The BERT algorithm is a bi-directional encoder algorithm, and when processing a word, the information of the word before and the word after the word can be considered, so that the semantics of the context can be obtained, character-level, word-level, sentence-level and even inter-sentence relationship features can be fully described, and the computer can be helped to better understand the language like a human being. By using the BERT algorithm, subtle differences among words can be better understood, and the obtained vector can better represent the true meaning of the common keywords.

And S308, calculating the similarity of the vector of the common keyword and the vector of the tag information of the plurality of pictures.

In step S308, the gallery has a plurality of pictures, the pictures in the gallery have labels, and the labels are text labels for the pictures. The picture provider can add a label to the picture and upload the picture to the gallery. Or, when receiving the picture, the gallery manager adds a label to the picture. A picture may have one or more labels, for example, a picture may be labeled with two labels, a "gift" and a "gift".

In the case that the picture has a tag, in step S308, tag information of the picture may be obtained, and embedded vector (Embedding) calculation is performed on the tag information of the picture by using the BERT algorithm, so as to obtain a vector of the tag information of the picture, which is used as a vector of the tag information of the picture. One feature of the embedded vector (Embedding) calculation is that it can be implemented to create a mapping from a high-dimensional vector to a low-dimensional vector, and the low-dimensional vector is used to characterize the picture, so that the calculation amount in step S308 can be reduced.

And calculating the similarity of the vector of the common key word of the plurality of small programs and the vector of the label information of the plurality of pictures, wherein the similarity of the vector of the common key word and the vector of the label information of the picture can reflect the association degree of the common key word and the picture.

In one embodiment, cosine similarity of the vector of the common keyword and the vector of the tag information of the picture is calculated. Cosine similarity is the measure of similarity between two vectors in vector space by measuring the cosine value of their included angle. The range of cosine values is between [ -1,1], the closer the cosine value is to 1, the closer the direction of the vector of the common key and the vector of the tag information of the picture is, and the higher the degree of association of the common key and the picture is. The closer the cosine value is to-1, the more opposite the direction of the vector indicating the common keyword and the vector of the tag information of the picture, the lower the degree of association of the common keyword and the picture.

In one embodiment, the distance between the vector of the common keyword and the vector of the tag information of the picture is calculated, and the closer the distance is, the higher the similarity of the two is. For example, the euclidean distance of the vector of the common keyword and the vector of the tag information of the picture is calculated. The smaller the Euclidean distance between the vector of the common keyword and the vector of the label information of the picture, the higher the similarity of the vector and the label information of the picture, and the higher the association degree between the common keyword and the picture. The larger the Euclidean distance between the vector of the common keyword and the vector of the label information of the picture is, the lower the similarity of the vector and the label information is, and the lower the association degree between the common keyword and the picture is.

S310, selecting pictures serving as image materials of the aggregation pages of the applets from the pictures according to the similarity.

In one embodiment, the picture with the highest similarity is configured as an icon or a background picture of an aggregate page of the plurality of applets. In a specific example, the picture with the highest similarity is configured as an icon of an aggregate page of the plurality of applets. In a specific example, the picture with the highest similarity is configured as a background picture of an aggregate page of the plurality of applets.

Referring to fig. 7 (a) -7 (c), the "payment" aggregation applet in fig. 7 (b) is an aggregation of the various specific payment applets in fig. 7 (c). Referring to fig. 7 (b), the icon of the "pay" aggregation applet is in the form of a rectangular frame, and the text "pay" is marked in the rectangular frame, which is not attractive enough and is not beneficial for the user to understand the purpose of the applet. By using the method of the embodiment of the specification, the common keyword of the specific payment applet aggregated by the payment aggregation applet is payment, and the lightning pattern marked with the payment is selected as the image material of the payment aggregation applet through similarity calculation of step S308. In the page 7 (a), when the user clicks the "applet", the jumped applet center page is shown in fig. 7 (d), in the page 7 (d), the icon of the "pay" aggregate applet is a lightning pattern marked with "fee", after the user sees the pattern, the user understands that the quick pay function can be realized through the aggregate applet, so that the user is attracted to use the pay applet, and the probability of using the pay applet by the user is improved.

In one embodiment, at least two pictures are selected from high to low according to the similarity, the selected pictures are fused to obtain a fused picture, and the fused picture is used as a picture material of an aggregation page of the plurality of applets. For example, pattern elements in the selected picture are extracted, and a fused picture is generated using these pattern elements. For example, the selected pictures are stitched together by a surf (Speeded Up Robust Features, accelerated robust feature) based image stitching algorithm to obtain a fused picture. In a specific example, the fused picture is configured as an icon of an aggregate page of the plurality of applets. In a specific example, the picture with the highest similarity is configured as a background picture of an aggregate page of the plurality of applets.

The methods of providing picture material for applets shown in fig. 2 and 3 may be implemented by the applet management platform shown in fig. 1.

The method for providing the picture material for the applet can accurately obtain the image material related to the applet, and the page can be beautified by utilizing the image material, so that a user can intuitively understand the purpose of the applet, and the applet can be attracted to the user.

The method for providing the picture material for the applet has high response speed and is suitable for various operation scenes.

The method for providing the picture material for the applet can reduce the labor work and even avoid manual intervention.

The method for providing the picture material for the applet can accurately obtain the image material related to the applet, the accuracy can reach more than 70%, and the labor work can be reduced by 80%.

The method for providing the picture material for the applet can accurately provide the related image material for the aggregated pages of a plurality of applets, and is particularly suitable for scenes needing to quickly build the aggregated pages.

< apparatus for providing Picture Material for applet >

< first embodiment >

Referring to fig. 4, this embodiment provides an apparatus 10 for providing picture material for applets, comprising the following modules:

a keyword extraction module 11, configured to obtain information of an applet, where the information of the applet includes a name, a category, and description information of the applet; mining keywords about the applet from the applet information;

the vectorization processing module 12 is configured to perform vectorization processing on the keyword to obtain a vector of the keyword;

A similarity calculation module 13, configured to calculate a similarity between the vector of the keyword and the vector of the tag information of the plurality of pictures;

and a selecting module 14, configured to select a picture as the image material of the applet from a plurality of pictures according to the similarity.

Optionally, the selecting module 14 is configured to configure the picture with the highest similarity as an icon or a background picture of the applet.

Optionally, the vectorization processing module 12 is configured to perform embedded vector calculation on the keyword through a BERT algorithm, so as to obtain a vector of the keyword.

Optionally, the method further comprises a picture vector extraction module. The picture vector extraction module is used for acquiring label information of the picture; and carrying out embedded vector calculation on the label information of the picture through a BERT algorithm to obtain a vector of the label information of the picture.

Alternatively, the similarity calculation module 13 calculates the similarity of the vector of the keyword and the vector of the tag information of the plurality of pictures, including: calculating cosine similarity of the vector of the keyword and the vector of the tag information of the plurality of pictures; or, calculating the Euclidean distance between the vector of the keyword and the vector of the tag information of the plurality of pictures.

< second embodiment >

Referring to fig. 5, this embodiment provides an apparatus 20 for providing picture material for applets, comprising the following modules:

a keyword extraction module 21, configured to obtain information of a plurality of applets, where the information of the applets includes names, categories and description information of the applets; mining common keywords about the plurality of applets from the information of the plurality of applets;

the vectorization processing module 22 is configured to perform vectorization processing on the common keywords to obtain vectors of the common keywords;

a similarity calculation module 23, configured to calculate a similarity between the vector of the common keyword and the vector of the tag information of the plurality of pictures;

and the selecting module 24 is configured to select, from a plurality of pictures, a picture that is an image material of an aggregate page of the plurality of applets according to the similarity.

Optionally, the selecting module 24 is configured to configure the picture with the highest similarity as an icon or a background picture of an aggregate page of the plurality of applets.

Optionally, the vectorization processing module 22 is configured to perform embedded vector calculation on the common keywords through a BERT algorithm, so as to obtain vectors of the common keywords.

Alternatively, the similarity calculation module 23 calculates the similarity between the vector of the common keyword and the vector of the tag information of the plurality of pictures, including: calculating cosine similarity of the vector of the common key word and the vector of the tag information of the plurality of pictures; or, calculating Euclidean distance between the vector of the common keyword and the vector of the tag information of the plurality of pictures.

< third embodiment >

Referring to fig. 6, this embodiment provides an apparatus 30 for providing picture material for an applet, comprising a processor 31 and a memory 32, the memory 32 storing a computer program which, when executed by the processor 31, implements the method for providing picture material for an applet of any of the previous embodiments.

The means for providing the picture material to the applet may be a server, the configuration of which may include, but is not limited to: processor, memory, interface device, communication device, input device, output device. The processor may include, but is not limited to, a central processing unit CPU, a microprocessor MCU, and the like. The memory may include, but is not limited to, ROM (read Only memory), RAM (random Access memory), nonvolatile memory such as a hard disk, and the like. The interface device may include, but is not limited to, a USB interface, a serial interface, a parallel interface, and the like. The communication means can for example perform wired or wireless communication, and may specifically comprise WiFi communication, bluetooth communication, 2G/3G/4G/5G communication, etc. Input devices 1015 may include, but are not limited to, a keyboard, mouse, and the like. The output device may include, but is not limited to, a display screen or the like. The configuration of the server may also include only a part of the devices described above.

The device for providing the picture material for the applet can accurately obtain the image material related to the applet.

The device for providing the picture material for the applet has high response speed and is suitable for various operation scenes.

The device for providing the picture material for the applet can reduce the labor work and even avoid manual intervention.

The device for providing the picture material for the applets can accurately provide the related image material for the aggregation pages of a plurality of applets.

< applet management platform >

The embodiment of the invention provides an applet management platform, which comprises any one of the devices for providing picture materials for applets. The specific configuration of the applet management platform can be seen in fig. 1.

< computer-readable Medium >

The present description also provides a computer-readable medium having stored thereon a computer program which, when executed by a processor, implements any of the foregoing methods of providing picture material for applets.

In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments. In particular, for the apparatus, device embodiments, since they are substantially similar to the method embodiments, the description is relatively simple, as relevant to see the section description of the method embodiments.

The foregoing describes specific embodiments of the present disclosure. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims can be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

Embodiments of the present description may be systems, methods, and/or computer program products. The computer program product may include a computer readable storage medium having computer readable program instructions embodied thereon for causing a processor to implement aspects of embodiments of the present description.

The computer readable storage medium may be a tangible device that can hold and store instructions for use by an instruction execution device. The computer readable storage medium may be, for example, but not limited to, an electronic storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of the computer-readable storage medium would include the following: portable computer disks, hard disks, random Access Memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static Random Access Memory (SRAM), portable compact disk read-only memory (CD-ROM), digital Versatile Disks (DVD), memory sticks, floppy disks, mechanical coding devices, punch cards or in-groove structures such as punch cards or grooves having instructions stored thereon, and any suitable combination of the foregoing. Computer-readable storage media, as used herein, are not to be construed as transitory signals per se, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (e.g., optical pulses through fiber optic cables), or electrical signals transmitted through wires.

The computer readable program instructions described herein may be downloaded from a computer readable storage medium to a respective computing/processing device or to an external computer or external storage device over a network, such as the internet, a local area network, a wide area network, and/or a wireless network. The network may include copper transmission cables, fiber optic transmissions, wireless transmissions, routers, firewalls, switches, gateway computers and/or edge servers. The network interface card or network interface in each computing/processing device receives computer readable program instructions from the network and forwards the computer readable program instructions for storage in a computer readable storage medium in the respective computing/processing device.

Computer program instructions for carrying out operations of embodiments of the present description may be assembly instructions, instruction Set Architecture (ISA) instructions, machine-related instructions, microcode, firmware instructions, state setting data, or source or object code written in any combination of one or more programming languages, including an object oriented programming language such as Smalltalk, c++ or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The computer readable program instructions may be executed entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any kind of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or may be connected to an external computer (for example, through the Internet using an Internet service provider). In some embodiments, aspects of embodiments of the present description are implemented by personalizing electronic circuitry, such as programmable logic circuitry, field Programmable Gate Arrays (FPGAs), or Programmable Logic Arrays (PLAs), with state information for computer-readable program instructions, which may execute the computer-readable program instructions.

Aspects of the present description embodiments are described herein with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the description. It will be understood that each block of the flowchart illustrations and/or block diagrams, and combinations of blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer-readable program instructions.

These computer readable program instructions may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions/acts specified in the flowchart and/or block diagram block or blocks. These computer readable program instructions may also be stored in a computer readable storage medium that can direct a computer, programmable data processing apparatus, and/or other devices to function in a particular manner, such that the computer readable medium having the instructions stored therein includes an article of manufacture including instructions which implement the function/act specified in the flowchart and/or block diagram block or blocks.

The computer readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other devices to cause a series of operational steps to be performed on the computer, other programmable apparatus or other devices to produce a computer implemented process such that the instructions which execute on the computer, other programmable apparatus or other devices implement the functions/acts specified in the flowchart and/or block diagram block or blocks.

The flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present description. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of instructions, which comprises one or more executable instructions for implementing the specified logical function(s). In some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions. It is well known to those skilled in the art that implementation by hardware, implementation by software, and implementation by a combination of software and hardware are all equivalent.

The embodiments of the present specification have been described above, and the above description is illustrative, not exhaustive, and not limited to the embodiments disclosed. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope of the various embodiments described. The terminology used herein was chosen in order to best explain the principles of the embodiments, the practical application, or the technical improvement of the technology in the marketplace, or to enable others of ordinary skill in the art to understand the embodiments disclosed herein.

Claims

1. A method of providing picture material for an applet, comprising the steps of:

vectorizing the common keywords to obtain vectors of the common keywords;

2. The method of claim 1, selecting a picture from a plurality of the pictures that is image material of an aggregate page of the plurality of applets according to the similarity, comprising:

3. The method of claim 1, the applet information further comprising information of a main page presentation of the applet.

4. The method of claim 1, vectorizing the common keywords to obtain vectors of the common keywords, comprising:

5. The method of claim 1, the vector of label information for the picture is obtained by:

acquiring label information of the picture;

6. The method of claim 1, calculating a similarity of the vector of the common keyword and the vector of tag information of a plurality of pictures, comprising:

7. An apparatus for providing picture material for an applet, comprising the following modules:

8. An apparatus for providing picture material for an applet, comprising a processor and a memory, said memory storing a computer program which, when executed by said processor, implements the method of any one of claims 1-6.