CN105335398B

CN105335398B - Service recommendation method and terminal

Info

Publication number: CN105335398B
Application number: CN201410344448.5A
Authority: CN
Inventors: 姜南
Original assignee: Huawei Technologies Co Ltd
Current assignee: Huawei Technologies Co Ltd
Priority date: 2014-07-18
Filing date: 2014-07-18
Publication date: 2020-08-25
Anticipated expiration: 2034-07-18
Also published as: CN105335398A

Abstract

The invention discloses a service recommendation method, which comprises the following steps: acquiring a short message in a terminal; carrying out language analysis on the short message, and identifying the potential intention of the user corresponding to the short message; and acquiring a service corresponding to the potential intention of the user according to the potential intention of the user. The embodiment of the invention also provides a corresponding terminal. The method can accurately identify the potential intention of the user corresponding to the short message by carrying out language analysis on the short message, and obtains the service corresponding to the potential intention of the user, so that the obtained service meets the requirements of the user.

Description

Service recommendation method and terminal

Technical Field

The invention relates to the technical field of information, in particular to a service recommendation method and a terminal.

Background

Short message applications on mobile terminals have become an indispensable application for users every day. Short Message applications are also diverse, such as traditional Short Message Service (SMS), Multimedia Short Message Service (MMS), and recently emerging WeChat, microblog, etc. The content of the short message is also wide, including communication among friends, mass-sending advertisements of enterprises, notification pushed by enterprises, and the like.

Short messages on mobile terminals cannot contain too much content due to word count limitations. But many of the content in the short message contains potential user requirements behind it. For example, potential product recommendation services, discounted information push services, may be involved behind the merchant name appearing in the short message; potential calendar services may be involved behind the time; the address back may relate to potential map navigation services, etc. If the potential needs of the user can be identified in the limited word number and the service and the advertisement are recommended in a targeted mode, the usability of the mobile terminal is improved, the stickiness of the user is improved, the targeting and the accuracy of service and advertisement pushing are improved, and the user's repugnance to the service and the advertisement is reduced.

The prior art provides a method for recommending services based on keywords in short messages. The method can comprise the following steps: the terminal identifies the keywords in the short message according to the keyword database; filtering out specific keywords related to the user from the recognized keywords according to the specific information of the user; highlighting the filtered keywords; after the user selects the keywords, selecting advertisements corresponding to the keywords selected by the user from an advertisement database according to the specific information of the user, and specifying context services corresponding to the keywords selected by the user; the selected advertisement and the specified contextual service are displayed.

In the process of research and practice of the prior art, the inventor of the present invention found that different short messages may contain the same keywords, but the intentions of users corresponding to different short messages may be completely different, so that the keywords in the short messages are identified only according to the keyword database, and advertisements or contextual services corresponding to the keywords are recommended to the users, which easily causes the recommended services not to match the intentions of the users, and the recommended services do not meet the real requirements of the users. For example, the short message content is "tomorrow leaves Sichuan university and ends life", the identified keyword is likely to be "Sichuan university" according to the keyword database, and a map service related to Sichuan university is recommended. However, as can be seen from the sentence, the "Sichuan" and "university" in the short message should be separated, and thus, the recommended Sichuan university-related map service basically does not meet the user's requirements. Also for example, the message content "with or without a hundred degree of interest developers ' congress", it is highly probable that "hundred degree" is identified for the prior art, and a "hundred degree" related search service or the like is recommended, which is actually a complete content from the "hundred degree developers ' congress" in a sentence pattern, and thus, the recommended "hundred degree" related search service basically does not meet the user's requirements.

Therefore, the identified keywords in the prior art cannot accurately reflect the intention of the short message, so that the recommended service does not meet the requirements of the user.

Disclosure of Invention

The invention aims to provide a service recommendation method and a terminal.

A first aspect of the present invention provides a service recommendation method, including:

acquiring a short message in a terminal;

carrying out language analysis on the short message, and identifying the potential intention of the user corresponding to the short message;

and acquiring a service corresponding to the potential intention of the user according to the potential intention of the user.

With reference to the first aspect, in a first implementation manner of the first aspect, the performing language parsing on the short message to obtain a potential intention of a user corresponding to the short message includes:

performing lexical analysis on the short message to acquire concepts and attributes of words in the short message and logic relations among the words;

determining a named entity corresponding to the word according to the concept or the attribute of the word;

and determining the potential intention of the user corresponding to the short message according to the named entity and the logical relation between the words.

With reference to the first implementation manner of the first aspect, in a second implementation manner of the first aspect, before determining, according to the named entity and the logical relationship between the words, the potential intention of the user corresponding to the short message, the method further includes:

determining the part of speech of the word, and identifying the syntactic structure of the short message according to the part of speech of the word;

the determining the potential intention of the user corresponding to the short message according to the named entity and the logical relationship between the words comprises:

and determining the potential intention of the user corresponding to the short message according to the named entity, the logical relation among the words and the syntactic structure of the short message.

With reference to the first aspect, or the first implementation manner of the first aspect, or the second implementation manner of the first aspect, in a third implementation manner of the first aspect, after performing language parsing on the short message to obtain a potential intention of a user corresponding to the short message, the method further includes:

when the short message corresponds to at least two potential intentions, acquiring the probability corresponding to each potential intention in the at least two potential intentions according to an intention model; and screening a preset number of potential intentions according to the probability corresponding to each potential intention in the at least two potential intentions.

With reference to the third implementation manner of the first aspect, in a fourth implementation manner of the first aspect, before the obtaining, according to a preset intention model, a probability corresponding to each potential intention of the at least two potential intentions, the method further includes:

and training the historical operation data of the user by using a machine learning method according to the historical operation data of the user to obtain the intention model.

With reference to the first aspect, or the first implementation manner of the first aspect, or the second implementation manner of the first aspect, or the third implementation manner of the first aspect, or the fourth implementation manner of the first aspect, in a fifth implementation manner of the first aspect, the obtaining, according to the potential intention of the user, a service corresponding to the potential intention of the user specifically includes:

determining parameters required for obtaining services corresponding to the potential intention of the user, and obtaining the context of the short message according to the required parameters;

generating a recommendation corresponding to the potential intent according to the context scenario.

With reference to the first aspect, or the first implementation manner of the first aspect, or the second implementation manner of the first aspect, or the third implementation manner of the first aspect, or the fourth implementation manner of the first aspect, or the fifth implementation manner of the first aspect, in a sixth implementation manner of the first aspect, after the acquiring a service corresponding to the potential intention of the user, the method further includes: when the number of the services is at least two, merging the services according to a preset service type to obtain a first service set;

and acquiring a preset number of services from the first service set, and displaying the preset number of services on an interface for a user to select after the preset number of services are sorted according to the fit degree with the short message.

A second aspect of the present invention provides a terminal, comprising:

a message acquiring unit, configured to acquire a short message in a terminal;

the intention identification unit is used for carrying out language analysis on the short message and identifying the potential intention of the user corresponding to the short message;

and the service acquisition unit is used for acquiring a service corresponding to the potential intention of the user according to the potential intention of the user.

With reference to the second aspect, in a first implementation manner of the second aspect, the intention identifying unit includes:

the lexical analysis subunit is used for carrying out lexical analysis on the short message to acquire concepts and attributes of the words in the short message and the logic relationship among the words;

the first determining subunit is used for determining a named entity corresponding to the term according to the concept or the attribute of the term;

and the second determining subunit is used for determining the potential intention of the user corresponding to the short message according to the named entity and the logical relationship between the words.

With reference to the first implementation manner of the second aspect, in a second implementation manner of the second aspect, the intention identifying unit further includes:

a syntax identification subunit, configured to determine a part of speech of the word, and identify a syntax structure of the short message according to the part of speech of the word;

the second determining subunit is specifically configured to,

With reference to the second aspect, or the first implementation manner of the second aspect, or the second implementation manner of the second aspect, in a third implementation manner of the second aspect, the terminal further includes:

a probability obtaining unit, configured to, when the short message obtained by the service obtaining unit corresponds to at least two potential intentions, obtain, according to an intention model, a probability corresponding to each potential intention of the at least two potential intentions;

and the intention filtering unit is used for screening a preset number of potential intentions according to the probability corresponding to each potential intention in the at least two potential intentions.

With reference to the third implementation manner of the second aspect, in a fourth implementation manner of the second aspect, the terminal further includes:

the model training unit is used for training the historical operation data of the user by utilizing a machine learning method according to the historical operation data of the user to obtain the intention model;

the probability obtaining unit is specifically configured to, when the short message obtained by the service obtaining unit corresponds to at least two potential intentions, obtain, according to the intention model trained by the model training unit, a probability corresponding to each potential intention of the at least two potential intentions.

With reference to the second aspect, or the first implementation manner of the second aspect, or the second implementation manner of the second aspect, or the third implementation manner of the second aspect, or the fourth implementation manner of the second aspect, in a fifth implementation manner of the second aspect, the service acquisition unit includes:

the context acquisition subunit is configured to determine parameters required to acquire a service corresponding to the potential intention of the user, and acquire a context of the short message according to the required parameters;

and the description generation subunit is used for generating a recommendation description corresponding to the potential intention according to the context scenario.

With reference to the second aspect, or the first implementation manner of the second aspect, or the second implementation manner of the second aspect, or the third implementation manner of the second aspect, or the fourth implementation manner of the second aspect, or the fifth implementation manner of the second aspect, in a sixth implementation manner of the second aspect, the terminal further includes:

a service merging unit, configured to merge the services according to a preset service type to obtain a first service set when the number of the services acquired by the service acquisition unit is at least two;

and the service display unit is used for acquiring a preset number of services from the first service set, sequencing the preset number of services according to the fit degree of the short message, and displaying the service on an interface for a user to select.

The method can accurately identify the potential intention of the user corresponding to the short message by performing language analysis on the short message, and obtain the service corresponding to the potential intention of the user, so that the obtained service meets the requirements of the user.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the description of the embodiments of the present invention will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings can be obtained by those skilled in the art.

FIG. 1 is a schematic diagram of an embodiment of a service recommendation method provided by the present invention;

FIG. 2 is a schematic diagram of another embodiment of a service recommendation method provided by the present invention;

FIG. 3 is a schematic diagram of one embodiment of a terminal provided by the present invention;

fig. 4 is a schematic diagram of another embodiment of the terminal provided in the present invention;

fig. 5 is a schematic diagram of another embodiment of the terminal provided by the present invention.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments that can be derived by one skilled in the art from the embodiments given herein are intended to be within the scope of the invention.

As shown in fig. 1, an embodiment of the present invention provides a service recommendation method, which can be applied to terminals such as a mobile phone and a tablet computer, and includes:

101. and acquiring the short message in the terminal.

The short message may be text information acquired by the terminal, and the text information includes, but is not limited to, a microblog, information of instant messaging software interaction, a short message, a notification, and the like.

102. And carrying out language analysis on the short message to obtain the potential intention of the user corresponding to the short message.

The language parsing may include lexical parsing, syntactic parsing. And carrying out language analysis on the short message to analyze the meaning to be expressed by the short message so as to obtain the potential intention of the user corresponding to the short message.

103. And acquiring a service corresponding to the potential intention of the user according to the potential intention of the user.

The service corresponding to the potential intention of the user may be a service provided on a network or a service provided by an application installed on a terminal.

According to the service recommendation method provided by the embodiment, the language of the short message is analyzed, so that the potential intention of the user corresponding to the short message can be accurately obtained, and the service corresponding to the potential intention of the user is obtained, so that the obtained service meets the requirements of the user.

In order to better understand the technical scheme of the invention, the following examples are provided for illustration in more detail.

As shown in fig. 2, an embodiment of the present invention further provides a method for recommending a service, which specifically includes:

201. acquiring a short message in a terminal;

the short message may be text information acquired by the terminal, and includes but is not limited to a microblog, information of instant messaging software interaction, a short message, a notification, and the like.

202. The lexical method analyzes the short message, obtains the concept and the attribute of the words in the short message, and determines the named entity corresponding to the words in the short message according to the concept or the attribute of the words;

specifically, one or more sentences in the short message are divided into one or more words, concepts, attributes and logical relations among the words in the short message are obtained, named entities corresponding to the words are determined according to the concepts or the attributes of the words, and the potential intentions of the user corresponding to the short message are determined according to the logical relations among the named entities and the words.

For example, the short message content "leave Sichuan in tomorrow", and the word segmentation result may output "leave" in tomorrow "and" Sichuan ". The named entity corresponding to the word in the short message is determined according to the concept, the attribute and the like of the word, and the result of the named entity can output 'tomorrow-time', 'Sichuan-place (or province)' and the like. The terminal can also standardize words in the short message according to the lexical analysis model and the word bank, and determine the corresponding relation between the words and the named entity. For example, "4 pm" in the short message "4 pm go to starbucking sitting" is normalized to xx month xx day of xx year (today date) 16: 00, the corresponding named entity is time, the corresponding named entity of starbucks is restaurant name, place name and music name, the corresponding named entity of sitting is dining event, and then the potential intention of the user corresponding to the short message is determined according to the logical relationship between the named entities and the words.

203. Determining the part of speech of words in the short message, and identifying the syntactic structure of the short message according to the part of speech of the words;

in a specific implementation, the short message may be segmented first, and the part of speech of each word in the short message is identified, where the part of speech includes a verb, a noun, an adjective, a quantifier, a pronoun, an adverb, and the like, so as to identify the syntactic structure of the short message according to the part of speech of the word. For example, the short message "go starbucks sit at 4 pm", "ROOT- > go (verb), go (schrench) - > 4 pm (time noun), go (object) - > starbucks (place noun), go (parallel relationship) - > sit".

204. Determining the potential intention of a user corresponding to the short message according to the named entity, the logical relation among the words and the syntactic structure of the short message;

it should be noted that, after the named entity corresponding to the word in the short message is identified in step 203, the potential intention of the user corresponding to the short message may be determined according to the logical relationship between the named entity and the word. If the potential intention of the user corresponding to the short message cannot be determined according to the logical relationship between the named entity and the word, after the syntactic structure of the short message is determined in step 204, the potential intention of the user corresponding to the short message is jointly determined according to the named entity, the logical relationship between the word and the syntactic structure of the short message.

205. Training the historical operation data of the user by using a machine learning method according to the historical operation data of the user to obtain an intention model;

in the embodiment, the machine learning method can be used for training the historical operation data of the user according to the historical operation data of the user and by combining the semantic structure of natural language, the society and the personal living habits of the user, so as to obtain the intention model. For example, the term "apple" may refer to fruit and may also refer to the brand name of the electronic product. When the short message in the terminal is "apple bought today", the user cannot determine whether the "apple" is a fruit or an electronic product by looking at the short message alone, and at this time, the user needs to determine the apple "by using a machine learning method according to the historical operation data of the user. For example, when "apple" in this sentence refers to fruit, words defining the weight of the fruit ("jin" ), the variety of the fruit (hong fu shi, jin guan), etc. may appear in the historical operation data of the user, and the real intention of the user can be determined according to these defining words; when "apple" in this language refers to an electronic product, a word for defining parameters such as a model, a configuration, and the like of the electronic product may appear in the historical operation data of the user. For example, the short message content "go to starbucks sitting bar in the afternoon", according to the historical operation data of the user, for example, when the word "sit and sit" appears in a common terminal, the user can search for a dining place, so that it can be known that the "sit and sit" is more inclined to the user to have a meal, and in addition, it can also be known that the "sit and sit" at this place refers to dining more than once by referring to natural language expression habits, so that the real intention of the user can be accurately known, and an intention model can be obtained.

206. According to the intention model, obtaining the probability corresponding to each potential intention, and obtaining a preset number of potential intentions according to the probability corresponding to each potential intention;

for example, when the content of the short message is "go starbucks sit at a bar in the afternoon", the potential intention of the user determined according to the steps 202 to 204 may be "get coupon of starbucks", "look up menu of starbucks", "query route of starbucks", etc., and according to the intention model, the user applies for the coupon of starbucks and sometimes looks up menu of starbucks every time the above-mentioned message appears on the terminal, then the probability corresponding to each potential intention will be: "recommend coupon-starbucks" > "recommend menu-starbucks" > "recommend route-starbucks".

If the number of potential intentions corresponding to one short message is large, only a preset number of potential intentions with high probability can be acquired according to the probability corresponding to each potential intention, so that the acquired potential intentions are closer to the real requirement of the user.

207. Obtaining a context scenario of the short message, and generating a recommendation explanation corresponding to the potential intention according to the context scenario;

after the potential intention of the user corresponding to the short message is determined, parameters required for obtaining the service corresponding to the potential intention of the user are determined, the context of the short message is obtained according to the required parameters, and the corresponding recommendation description is generated according to the context of the short message. For example, when the service recommended for the user is a route to a certain place, it is necessary to determine parameters required for recommending the service for the user, and in the present example, the parameters include at least a departure place and a destination. In this embodiment, the context of the short message may be a key word in the short message, may also be other user information related to the content of the short message, and may also be information for prompting the user to input. For example, when the short message is "6 pm at starbucks gate wait me", the determined potential intentions of the user are "set reminder-6 pm at starbucks gate wait xx", "query starbucks for a line". For the potential intention "set reminder-6 pm at the gate of starbucks, etc. xx", the key words "6 pm", "starbucks", "etc. xx" in the short message can be extracted, and then the recommendation "set xx year xx month xx for you and 6 pm at the gate of starbucks, etc. xx event reminder corresponding to the potential intention" is generated by combining the time information of the day, such as the year, month, day, etc. displayed by the terminal; for the potential intention "inquire about the route to starbucks", the context scenario of the short message may be the current location information xx of the user, which may be obtained from the GPS location of the user terminal, or from information input by the user according to the prompt, and the generated recommendation corresponding to the potential intention is "inquire about the route from xx to starbucks" for you.

In addition, when the context scenario of the short message is other user information related to the content of the short message, the user information may also be user data stored in a terminal memory or storage device, including but not limited to other short messages, address books, memos, reminders, photos, applications, video, audio, mail, bookmarks, web browsing records, purchase records of goods/services, hotel reservations, airline ticket purchase records, some preference settings of the user, etc., in addition, the user information may also be related information of the terminal used by the user, such as hardware information and software information of the terminal, including but not limited to date and clock information, location information (e.g., GPS, country, city, etc.), information generated by the sensors, the operating system of the terminal and the software running on the terminal, the status and events of processes, services, etc.

208. Acquiring a service corresponding to the potential intention of the user according to the recommendation selected by the user;

in this embodiment, the key words used for generating the recommendation descriptions in the short message may be specially displayed, each key word corresponds to one or more potential intentions, each potential intention has a corresponding recommendation description, and when a user focuses on a certain key word, the corresponding recommendation description is displayed for the user.

For example, for the short message "am 6 am at starbucks gate, etc." specially displaying the key words in the short message may be "6 amIn thatStar BakerThe entrance and the like ", wherein" 6 pm "corresponds to the potential intention of the user" set reminder "," starbucks "corresponds to the potential intention of the user" inquire about route to starbucks "," inquire about menu of starbucks "," inquire about coupon of starbucks ", and the like. The terminal can monitor the user behavior, and when the situation that the user clicks a preset certain key word such as 'starbucks' is monitored, the corresponding key word is displayed for the userThe recommended instructions of (1). Specifically, in the present example, when the user clicks "starbucks", recommendation descriptions corresponding to the three potential intentions are displayed for the user, and the display of the recommendation descriptions may be displayed in sequence according to the probability corresponding to each intention.

And when the user selects a certain recommended explanation, acquiring a service corresponding to the potential intention corresponding to the recommended explanation.

The service may be some applications that call the terminal itself, such as a reminder application, an alarm application, or a memo application on the terminal, or may be obtained from a third party, such as a server, other terminals, and the like.

Of course, in the above example, instead of selecting the recommended specification by the user, the service corresponding to all the recommended specifications may be automatically acquired after the recommended specification stays on the terminal display interface for a preset time.

209. The acquired services are combined to obtain a first service set, a preset number of services are acquired from the first service set, and the preset number of services are displayed on an interface for a user to select.

After the service corresponding to the potential intention of the user is obtained, the service contents can be merged to obtain a first service set, the purpose of doing so is to remove repeated services, and then the services in the first service set can be sorted according to the fit degree of the short message and then displayed on an interface of a terminal for the user to select.

The degree of the service to the short message may be the degree of the service to the key word in the short message, for example, when the short message is "eat at night to near the upper place", there is a key word "near the upper place", and the service corresponding to the potential intention of the user may have restaurants near the upper place recommended for the user, the closer the restaurant in the distance to the short message is, the higher the degree of the short message to the restaurant, and therefore, the services may be ranked in order of the degree of the short message to be displayed on the terminal interface, so as to facilitate the user to select. In addition, the degree of compliance of the service with the short message may also include, but is not limited to, the condition importance degree of the unified recognition judgment formed in daily life, for example, in the recommended coupon service, "preferential price" may be more preferable than "expiration time", and the degree of compliance of the service with the short message may also be judged according to the habit and history data of the user.

In the embodiment, the potential intention of the user is determined through language analysis, the recommendation explanation corresponding to the potential intention is generated to be selected by the user, and then the corresponding service is provided for the user, so that the flexibility of service recommendation is improved, the service recommended for the user is the service which best meets the potential needs of the user, and the user can conveniently select the service meeting the requirements.

Referring to fig. 3, a terminal 300 according to an embodiment of the present invention is described below, where the terminal is configured to perform all the methods stated above, and includes:

a message acquiring unit 301, configured to acquire a short message in a terminal;

an intention identifying unit 302, configured to perform language parsing on the short message acquired by the message acquiring unit 301 to acquire a potential intention of the user corresponding to the short message;

a service obtaining unit 303, configured to obtain a service corresponding to the potential intention of the user according to the potential intention of the user obtained by the intention identifying unit 302.

In this embodiment, the intention identifying unit may accurately identify the potential intention of the user corresponding to the short message by performing language parsing on the short message, and the service acquiring unit may acquire the service corresponding to the potential intention of the user, so that the acquired service may meet the requirement of the user.

For easy understanding, the following further describes the terminal according to the embodiment of the present invention, and referring to fig. 4, the terminal 400 according to the embodiment includes:

a message acquiring unit 401, configured to acquire a short message in a terminal;

an intention identifying unit 402, configured to perform language parsing on the short message acquired by the message acquiring unit 401, and acquire a potential intention of the user corresponding to the short message;

a model training unit 403, configured to train historical operation data of a user by using a machine learning method according to the historical operation data of the user, and obtain an intention model;

a probability obtaining unit 404, configured to, when the short message corresponds to at least two potential intentions, obtain a probability corresponding to each potential intention of the at least two potential intentions according to the intention model trained by the model training unit 403;

an intention filtering unit 405, configured to filter a preset number of potential intentions according to the probability corresponding to each potential intention in the at least two potential intentions acquired by the probability acquiring unit 404;

a service obtaining unit 406, configured to obtain, according to the potential intention of the user, a service corresponding to the potential intention of the user;

a service merging unit 407, configured to merge the services according to a preset service type when the number of the services is at least two, to obtain a first service set;

the service display unit 408 is configured to obtain a preset number of services from the first service set, and display the preset number of services on an interface for a user to select after the preset number of services are sorted according to the degree of attachment to the short message.

In addition, the intention identifying unit 402 specifically includes:

a lexical analysis subunit 4021, configured to perform lexical analysis on the short message, and acquire concepts and attributes of words in the short message and a logical relationship between the words;

a first determining subunit 4022, configured to determine a named entity corresponding to the term according to the concept or attribute of the term;

a syntax identifying subunit 4023, configured to determine a part of speech of the word, and identify a syntax structure of the short message according to the part of speech of the word;

a second determining subunit 4024, configured to determine a potential intention of the user corresponding to the short message according to the named entity, the logical relationship between the words, and the syntactic structure of the short message.

The service obtaining unit 406 specifically includes:

a context obtaining subunit 4061, configured to determine parameters required to obtain a service corresponding to the potential intention of the user, and obtain a context of the short message according to the required parameters;

an explanation generating sub-unit 4062 is configured to generate a recommended explanation corresponding to the potential intention according to the context scenario.

For further understanding, the following describes the interaction manner between the units of the terminal 400 in this embodiment in a practical application scenario:

first, the message obtaining unit 401 obtains a short message in the terminal, where the short message may be text information obtained by the terminal, and includes but is not limited to a microblog, information of instant messaging software interaction, a short message, a notification, and the like.

The lexical analysis subunit 4021 divides one or more sentences in the short message into one or more words, obtains concepts and attributes of the words in the short message and a logical relationship between the words, and the first determination subunit 4022 determines a named entity corresponding to the word according to the concepts or attributes of the words and determines a potential intention of a user corresponding to the short message according to the logical relationship between the named entity and the word.

For example, the content of the short message is "leave from Sichuan in tomorrow", and the result of word segmentation by the lexical analysis subunit 4021 can output "leave from Sichuan in tomorrow". The first determining subunit 4022 determines a named entity corresponding to a word in the short message according to the concept, attribute, and the like of the word, and the result of the named entity may output "tomorrow-time", "sichuan-place (or province)", and the like. In addition, the lexical analysis subunit 4021 may normalize words in the short message according to a lexical analysis model and a lexicon, and determine a correspondence between the words and the named entity. For example, "4 pm" in the short message "4 pm go to starbucking sitting" is normalized to xx month xx day of xx year (today date) 16: 00, the corresponding named entity is "time", "starbucks" the corresponding named entity is "restaurant name", "place name", "music name", and "sit" the corresponding named entity is "dining event", and then the second determining subunit 4024 determines the potential intention of the user corresponding to the short message according to the logical relationship between the named entities and the words.

If the second determining subunit 4024 cannot determine the potential intention of the user corresponding to the short message according to the logical relationship between the named entity and the word, the syntax identifying subunit 4023 determines the part of speech of the word in the short message, and identifies the syntax structure of the short message according to the part of speech of the word. Specifically, the syntax identifying subunit 4023 may perform word segmentation on the short message, and identify the part of speech of each word in the short message, where the part of speech includes a verb, a noun, an adjective, a quantifier, a pronoun, an adverb, and the like, so as to identify the syntax structure of the short message according to the part of speech of the word. For example, the short message "go starbucks sit at 4 pm", "ROOT- > go (verb), go (schrench) - > 4 pm (time noun), go (object) - > starbucks (place noun), go (parallel relationship) - > sit".

The second determining subunit 4024 determines the potential intention of the user corresponding to the short message jointly according to the named entity determined by the first determining subunit 4022, the logical relationship between the words parsed by the lexical parsing subunit 4021, and the syntactic structure of the short message recognized by the syntactic recognition subunit 4023.

After the second determining subunit 4024 determines the potential intention of the user, the model training unit 403 trains the historical operation data of the user by using a machine learning method according to the historical operation data of the user to obtain an intention model. In a specific implementation, the model training unit 403 may train the historical operation data of the user by using a machine learning method according to the historical operation data of the user and by combining the semantic structure of natural language, the society, and the personal habits of the user, so as to obtain an intention model. For example, the term "apple" may refer to fruit and may also refer to the brand name of the electronic product. When the short message in the terminal is "apple bought today", the user cannot determine whether the "apple" is a fruit or an electronic product by looking at the short message alone, and at this time, the user needs to determine the apple "by using a machine learning method according to the historical operation data of the user. For example, when "apple" in this sentence refers to fruit, in the historical operation data of the user, words that define the weight of the fruit ("jin" ), the variety of the fruit (hong fu shi, jin guan), and the like may appear; when "apple" in this language refers to an electronic product, a word for defining parameters such as a model, a configuration, and the like of the electronic product may appear in the historical operation data of the user. For example, the short message content "go to starbucks sitting bar in the afternoon", according to the historical operation data of the user, for example, when the word "sit and sit" appears in a common terminal, the user can search for a dining place, so that it can be known that the "sit and sit" is more inclined to the user to have a meal, and in addition, it can also be known that the "sit and sit" at this place refers to dining more than once by referring to natural language expression habits, so that the real intention of the user can be accurately known, and an intention model can be obtained.

The probability obtaining unit 404 obtains a probability corresponding to each potential intention according to the intention model trained by the model training unit 403, and obtains a preset number of potential intentions according to the probability corresponding to each potential intention. For example, when the content of the short message is "going to starbucks sitting bar in the afternoon", the potential intention of the user determined by the second determining subunit 4024 may be "obtaining a coupon of starbucks", "viewing a menu of starbucks", "querying a route to starbucks", etc., and according to the intention model, every time the above-mentioned message appears on the terminal, the user applies for the coupon of starbucks, sometimes also viewing the menu of starbucks, and then the probability corresponding to each potential intention will be: "recommend coupon-starbucks" > "recommend menu-starbucks" > "query map-starbucks".

If there are a large number of potential intentions corresponding to a short message, the intention filtering unit 405 may only obtain a preset number of potential intentions with a high probability according to the probability corresponding to each potential intention, so as to make the obtained potential intentions closer to the real requirement of the user.

Next, the scenario acquisition sub-unit 4061 determines parameters required to acquire a service corresponding to the potential intention of the user, acquires a context scenario of the short message according to the required parameters, and generates a recommendation corresponding to the potential intention according to the context scenario. For example, when the service recommended for the user is a route to a certain place, it is necessary to determine parameters required for recommending the service for the user, and in the present example, the parameters include at least a departure place and a destination. In this embodiment, the context of the short message may be a key word in the short message, may also be other user information related to the content of the short message, and may also be information for prompting the user to input. For example, when the short message is "6 pm at starbucks gate wait me", the determined potential intentions of the user are "set reminder-6 pm at starbucks gate wait xx", "query starbucks for a line". For the potential intention "set reminder-6 pm at starbucks gate etc xx", the scenario acquisition subunit 4061 may extract the key words "6 pm", "starbucks", "etc xx" in the short message, and then the description generation subunit 4062 generates a recommendation description "set up xx year xx month xx afternoon 6 pm at starbucks etc xx event reminder" for you in combination with the time information of the day, such as year, month, day, etc. displayed by the terminal itself, corresponding to the potential intention "; for the potential intention "inquire about the route to starbucks", the context scenario of the short message acquired by the scenario acquisition sub-unit 4061 may be the current location information xx of the user, and the location information may be acquired by GPS positioning of the user terminal, or may be acquired from information input by the user according to a prompt, and the recommendation generated by the description generation sub-unit 4062 and corresponding to the potential intention is "inquire about the route from xx to starbucks" for you.

In this embodiment, the description generating subunit 4062 may specifically display the key terms used for generating the recommendation description in the short message, where each key term corresponds to one or more potential intentions, each potential intention has a corresponding recommendation description, and when a user focuses on a certain key term, the corresponding recommendation description is displayed for the user.

For example, for the short message "i am at starbucks gate at 6 pm", it may be that the description generation subunit 4062 displays the key words in the short message in a special way "6 amIn thatStar BakerThe entrance and the like ", wherein" 6 pm "corresponds to the potential intention of the user" set reminder "," starbucks "corresponds to the potential intention of the user" inquire about route to starbucks "," inquire about menu of starbucks "," inquire about coupon of starbucks ", and the like. The description generation subunit 4062 may monitor the behavior of the user, and when it is monitored that the user clicks a preset certain key word, for example, "starbucks", a corresponding recommendation description is displayed for the user. Specifically, in the present example, when the user clicks "starbucks", recommendation descriptions corresponding to the three potential intentions are displayed for the user, and the display of the recommendation descriptions may be displayed in sequence according to the probability corresponding to each intention.

When the user selects a certain recommended specification, the service obtaining unit 406 obtains a service corresponding to the potential intention corresponding to the recommended specification.

The service acquired by the service acquiring unit 406 may be some application that calls the terminal itself, such as a reminder application, an alarm application, or a memo application on the terminal, or may be a service acquired from a third party, such as a server, other terminal, or the like.

Of course, in the above example, instead of waiting for the user to select the recommended specification, the service obtaining unit 406 may automatically obtain the services corresponding to all the recommended specifications after the recommended specification stays on the terminal display interface for a preset time.

After the service acquiring unit 406 acquires the corresponding service, the service merging unit 407 merges the acquired services to obtain a first service set, which is to remove the duplicated services, and then the service displaying unit 408 may sort the services in the first service set according to the degree of fit with the short message, and display the sorted services on the interface of the terminal for the user to select.

The degree of the service's attachment to the short message may be the degree of the service's attachment to a key word in the short message, for example, when the short message is "eat at night to the vicinity of the upper place", there is a key word "vicinity of the upper place", and the service corresponding to the potential intention of the user may have restaurants in the vicinity of the upper place recommended to the user, the closer restaurants in distance have higher attachment to the short message, and therefore, the service display unit 408 may display the services on the terminal interface in order of the degree of the attachment to the short message from high to low, so as to facilitate the user's selection. In addition, the degree of compliance of the service with the short message may also include, but is not limited to, the condition importance degree of the unified recognition judgment formed in daily life, for example, in the recommended coupon service, "preferential price" may be more preferable than "expiration time", and the degree of compliance of the service with the short message may also be judged according to the habit and history data of the user.

In the embodiment, the intention identification unit determines the potential intention of the user through language analysis, the explanation generation unit generates the recommendation explanation corresponding to the potential intention for the user to select, and then the service acquisition unit provides the corresponding service for the user, so that the flexibility of service recommendation is improved, the service recommended for the user is the service which best meets the potential needs of the user, and the user can conveniently select the service which meets the needs.

Referring to fig. 5, fig. 5 is a schematic view of another embodiment of the terminal of the present invention, the terminal 500 of the present embodiment may be used to implement the service recommendation method provided by the above embodiment, and in practical applications, the terminal 500 may be an electronic device such as a mobile phone and a notebook computer. Specifically, the method comprises the following steps:

the terminal 500 may include RF (Radio Frequency) circuitry 510, memory 520 including one or more computer-readable storage media, an input unit 530, a display unit 540, a sensor 550, audio circuitry 560, a WiFi (wireless fidelity) module 570, a processor 580 including one or more processing cores, and a power supply 590. Those skilled in the art will appreciate that the configuration shown in fig. 5 is not intended to be limiting of terminal 500 and may include more or fewer components than those shown, or some components may be combined, or a different arrangement of components. Wherein:

RF circuitry 510 may be used for receiving and transmitting signals during messaging or telephony, and in particular, for receiving downlink information from a base station and processing the received downlink information by one or more processors 580; in addition, data relating to uplink is transmitted to the base station. In general, RF circuitry 510 includes, but is not limited to, an antenna, at least one Amplifier, a tuner, one or more oscillators, a Subscriber Identity Module (SIM) card, a transceiver, a coupler, an LNA (Low Noise Amplifier), a duplexer, and the like. In addition, RF circuit 510 may also communicate with networks and other devices via wireless communication. The wireless communication may use any communication standard or protocol, including but not limited to GSM (Global System for Mobile communications), GPRS (General Packet Radio Service), CDMA (Code Division Multiple Access), WCDMA (Wideband Code Division Multiple Access), LTE (Long Term Evolution), e-mail, SMS (short messaging Service), etc.

The memory 520 may be used to store software programs and modules, and the processor 580 executes various functional applications and data processing by operating the software programs and modules stored in the memory 520. The memory 520 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application program required by at least one function (such as a sound playing function, an image playing function, etc.), and the like; the storage data area may store data created according to the use of the storage device (such as audio data, a phonebook, etc.). Further, the memory 520 may include high speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other volatile solid state storage device. Accordingly, the memory 520 may also include a memory controller to provide the processor 580 and the input unit 530 access to the memory 520.

The input unit 530 may be used to receive input numeric or character information and generate a keyboard, mouse, joystick, optical or trackball signal input related to user setting and function control. In particular, the input unit 530 may include a touch sensitive surface 531 as well as other input devices 532. The touch sensitive surface 531, also referred to as a touch display screen or a touch pad, may collect touch operations by a user on or near the touch sensitive surface 531 (e.g. operations by a user on or near the touch sensitive surface 531 using a finger, a stylus, or any other suitable object or attachment) and drive the corresponding connection device according to a predetermined program. Alternatively, the touch sensitive surface 531 may comprise two parts, a touch detection means and a touch controller. The touch detection device detects the touch direction of a user, detects a signal brought by touch operation and transmits the signal to the touch controller; the touch controller receives touch information from the touch sensing device, converts the touch information into touch point coordinates, and sends the touch point coordinates to the processor 580, and can receive and execute commands sent by the processor 580. In addition, the touch sensitive surface 531 may be implemented using various types of resistive, capacitive, infrared, and surface acoustic waves. The input unit 530 may comprise other input devices 532 in addition to the touch sensitive surface 531. In particular, other input devices 532 may include, but are not limited to, one or more of a physical keyboard, function keys (such as volume control keys, switch keys, etc.), a trackball, a mouse, a joystick, and the like.

The display unit 540 may be used to display information input by or provided to a user and various graphic user interfaces of the terminal, which may be configured by graphics, text, icons, video, and any combination thereof. The Display unit 540 may include a Display panel 541, and optionally, the Display panel 541 may be configured in the form of an LCD (Liquid Crystal Display), an OLED (Organic Light-Emitting Diode), or the like. Further, the touch-sensitive surface 531 can overlie the display panel 541 such that, when a touch event is detected at or near the touch-sensitive surface 531, it is passed to the processor 580 for determining the type of touch event, whereupon the processor 580 provides a corresponding visual output on the display panel 541 in dependence upon the type of touch event. Although in FIG. 5 the touch sensitive surface 531 and the display panel 541 are shown as two separate components to implement input and output functions, in some embodiments the touch sensitive surface 531 and the display panel 541 may be integrated to implement input and output functions.

The terminal 500 can also include at least one sensor 550, such as a light sensor, motion sensor, and other sensors. Specifically, the light sensor may include an ambient light sensor that adjusts the brightness of the display panel 541 according to the brightness of ambient light, and a proximity sensor that turns off the display panel 541 and/or a backlight when the terminal 500 moves to the ear. As one of the motion sensors, the gravity acceleration sensor can detect the magnitude of acceleration in each direction (generally, three axes), detect the magnitude and direction of gravity when the terminal is stationary, and can be used for applications of recognizing terminal gestures (such as horizontal and vertical screen switching, related games, magnetometer gesture calibration), vibration recognition related functions (such as pedometer and tapping), and the like; as for other sensors such as a gyroscope, a barometer, a hygrometer, a thermometer, and an infrared sensor, which can be configured in the terminal 500, detailed descriptions thereof are omitted.

Audio circuitry 560, speaker 561, microphone 562 may provide an audio interface between a user and the device. The audio circuit 560 may transmit the electrical signal converted from the received audio data to the speaker 561, and convert the electrical signal into a sound signal by the speaker 561 for output; on the other hand, the microphone 562 converts the collected sound signal into an electric signal, is received by the audio circuit 560 and converted into audio data, and then outputs the audio data to the processor 580 for processing, and then to the RF circuit 510 for transmission to, for example, another terminal, or outputs the audio data to the memory 520 for further processing. The audio circuit 560 may also include an earbud jack to provide communication of a peripheral headset with the terminal.

WiFi belongs to a short-distance wireless transmission technology, and the terminal 500 may help a user send and receive e-mails, browse web pages, access streaming media, and the like through the WiFi module 570, and provide the user with wireless broadband internet access. Although fig. 5 shows the WiFi module 570, it is understood that it does not belong to the essential constitution of the device, and may be omitted entirely as needed within the scope not changing the essence of the invention.

The processor 580 is a control center of the device that predicts the user's off-network, connects various parts of the entire device using various interfaces and lines, and performs various functions of the storage device and processes data by running or executing software programs and/or modules stored in the memory 520 and calling up data stored in the memory 520, thereby monitoring the storage device as a whole. Optionally, processor 580 may include one or more processing cores; optionally, processor 580 may integrate an application processor, which handles primarily the operating system, user interface, applications, etc., and a modem processor, which handles primarily the wireless communications. It will be appreciated that the modem processor described above may not be integrated into processor 580.

Terminal 500 also includes a power supply 590 (e.g., a battery) for powering the various components, which may optionally be logically connected to processor 580 via a power management system that may enable managing charging, discharging, and power consumption management functions. The power supply 590 may also include one or more dc or ac power sources, recharging systems, power failure detection circuitry, power converters or inverters, power status indicators, and any other components.

Although not shown, the terminal 500 may further include a camera, a bluetooth module, etc., which will not be described herein. In particular, in this embodiment, the terminal 500 includes a memory 520, and one or more programs, wherein the one or more programs are stored in the memory 520 and configured to be executed by the one or more processors 580, the one or more programs including instructions for:

acquiring a short message in a terminal;

carrying out language analysis on the short message to obtain the potential intention of the user corresponding to the short message;

Optionally, performing language parsing on the short message to obtain a potential intention of the user corresponding to the short message, including:

Optionally, before determining the potential intention of the user corresponding to the short message according to the named entity and the logical relationship between the words, the method further includes:

Optionally, after performing language parsing on the short message and obtaining the potential intention of the user corresponding to the short message, the method further includes:

when the short message corresponds to at least two potential intentions, acquiring the probability corresponding to each potential intention in the at least two potential intentions according to an intention model;

and screening a preset number of potential intentions according to the probability corresponding to each potential intention in the at least two potential intentions.

Optionally, before obtaining the probability corresponding to each potential intention of the at least two potential intentions according to the preset intention model, the method further includes:

Optionally, the obtaining, according to the potential intention of the user, a service corresponding to the potential intention of the user specifically includes:

Optionally, after acquiring the service corresponding to the potential intention of the user, the method further includes:

when the number of the services is at least two, merging the services according to a preset service type to obtain a first service set;

It should be noted that the terminal 500 provided in the embodiment of the present invention may also be used to implement other functions in the foregoing device embodiments, and details are not described herein again.

In the embodiments provided in the present application, it should be understood that the disclosed apparatus may be implemented in other manners. For example, the above-described embodiments of the apparatus are merely illustrative, and for example, the division of the units is only one type of division of logical functions, and there may be other divisions when actually implementing, for example, a plurality of units or components may be combined or may be integrated into another system, or some features may be omitted, or not implemented. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection of some interfaces, devices or units, and may be an electric or other form.

The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.

In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.

The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a removable hard disk, a magnetic or optical disk, and other various media capable of storing program codes.

The above-mentioned embodiments are only used for illustrating the technical solutions of the present invention, and not for limiting the same; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.

Claims

1. A service recommendation method, comprising:

acquiring a short message in a terminal;

performing language analysis on the short message to obtain the potential intention of a user corresponding to the short message, wherein the language analysis on the short message can obtain one or more words in the short message; the language parsing of the short message to obtain the potential intention of the user corresponding to the short message includes: performing lexical analysis on the short message to acquire concepts and attributes of words in the short message and logic relations among the words; determining a named entity corresponding to the word according to the concept or the attribute of the word; determining the potential intention of the user corresponding to the short message according to the named entity and the logic relation between the words;

when the short message corresponds to the potential intentions of at least two users, acquiring the potential intentions of a preset number of users according to an intention model, wherein the intention model is obtained by training based on historical operation data of the users, and the historical operation data of the users are operation data of the users aiming at the one or more words in history; and acquiring services corresponding to the potential intentions of the users in the preset number according to the potential intentions of the users in the preset number.

2. The method of claim 1, wherein before determining the potential intent of the user corresponding to the short message according to the named entity and the logical relationship between the words, the method further comprises:

3. The method of any one of claims 1 to 2, wherein the obtaining of the potential intentions of the preset number of users according to the intention model comprises:

according to the intention model, obtaining the probability corresponding to each potential intention in the potential intentions of the at least two users;

screening the potential intentions of the users in the preset number according to the probability corresponding to each potential intention in the potential intentions of the at least two users.

4. The method according to claim 3, wherein before obtaining the probability corresponding to each potential intention of the at least two potential intentions according to a preset intention model, the method further comprises:

5. The method according to claim 4, wherein the obtaining, according to the potential intention of the user, the service corresponding to the potential intention of the user specifically comprises:

6. The method of claim 5, wherein after obtaining the service corresponding to the potential intent of the user, further comprising:

7. A terminal, comprising:

a message acquiring unit, configured to acquire a short message in a terminal;

the intention identification unit is used for carrying out language analysis on the short message to obtain the potential intention of the user corresponding to the short message, wherein the intention identification unit can carry out language analysis on the short message to obtain one or more words in the short message; the intention identifying unit further includes: the lexical analysis subunit is used for carrying out lexical analysis on the short message to acquire concepts and attributes of the words in the short message and the logic relationship among the words; the first determining subunit is used for determining a named entity corresponding to the term according to the concept or the attribute of the term; the second determining subunit is used for determining the potential intention of the user corresponding to the short message according to the named entity and the logical relationship between the words;

when the short message corresponds to the potential intentions of at least two users, acquiring the potential intentions of a preset number of users according to an intention model, wherein the intention model is obtained by training based on historical operation data of the users, and the historical operation data of the users are operation data of the users aiming at the one or more words in history;

and the service acquisition unit is used for acquiring the services corresponding to the potential intentions of the preset number of users according to the potential intentions of the preset number of users.

8. The terminal of claim 7, wherein the intention identifying unit further comprises:

the second determining subunit is specifically configured to,

9. The terminal according to any of claims 7 to 8, characterized in that the terminal further comprises:

a probability obtaining unit, configured to obtain, according to the intention model, a probability corresponding to each potential intention in the potential intentions of the at least two users;

and the intention filtering unit is used for screening the potential intentions of the preset number of users according to the probability corresponding to each potential intention in the potential intentions of the at least two users.

10. The terminal of claim 9, wherein the terminal further comprises:

11. The terminal of claim 10, wherein the service acquisition unit comprises:

12. The terminal of claim 11, wherein the terminal further comprises: