CN106326338B - Service providing method and device based on search engine - Google Patents
Service providing method and device based on search engine Download PDFInfo
- Publication number
- CN106326338B CN106326338B CN201610630338.4A CN201610630338A CN106326338B CN 106326338 B CN106326338 B CN 106326338B CN 201610630338 A CN201610630338 A CN 201610630338A CN 106326338 B CN106326338 B CN 106326338B
- Authority
- CN
- China
- Prior art keywords
- search
- service
- search request
- vector
- user
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Evolutionary Computation (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- Artificial Intelligence (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a service providing method and a device based on a search engine, wherein the method comprises the following steps: judging whether a search request input to a search engine by a user contains a service requirement or not according to a pre-trained classifier; if the search request contains the service requirement, acquiring a search characteristic vector of the search request; acquiring a service vector corresponding to the search feature vector according to pre-trained service resource marking data; and providing the connection service information corresponding to the search request to the user through the search engine according to the service vector. The method excavates the service requirement of the user by deeply analyzing the search request of the user, provides related connection service information for the user, strengthens the function of a search engine and enables the search result to better meet the requirement of the user.
Description
Technical Field
The invention relates to the technical field of information processing, in particular to a service providing method and device based on a search engine.
Background
With the development of internet technology, search engines are seen as the main way to retrieve network information. With the penetration of the internet into traditional industry, especially third industry services, the search requirements of users are more and more diversified.
However, when the related search engine provides a search service for the user, the character string input by the user is required to be strictly matched with the search result, the returned search result is the same as the search request input by the user, and the search result provided for the user is limited.
Disclosure of Invention
The object of the present invention is to solve at least to some extent one of the above mentioned technical problems.
Therefore, an object of the present invention is to provide a service providing method based on a search engine, which exploits the service requirements of a user by deeply analyzing the search request of the user, and provides related connection service information for the user, thereby enhancing the function of the search engine, and enabling the search results to better satisfy the user requirements.
The second purpose of the invention is to provide a service providing device based on a search engine.
In order to achieve the above object, an embodiment of a first aspect of the present invention provides a service providing method based on a search engine, including:
judging whether a search request input to a search engine by a user contains a service requirement or not according to a pre-trained classifier;
if the search request contains service requirements, acquiring a search characteristic vector of the search request;
acquiring a service vector corresponding to the search feature vector according to pre-trained service resource marking data;
and providing connection service information corresponding to the search request to the user through the search engine according to the service vector.
The service providing method based on the search engine judges whether a search request input to the search engine by a user contains a service requirement according to a pre-trained classifier, if the search request contains the service requirement, a search characteristic vector of the search request is obtained, a service vector corresponding to the search characteristic vector is obtained according to pre-trained service resource marking data, and then connection service information corresponding to the search request is provided to the user through the search engine according to the service vector. Therefore, the service requirements of the users are mined by deeply analyzing the search requests of the users, and the related connection service information is provided for the users in the search engine, so that the function of the search engine is strengthened, and the search results can better meet the requirements of the users.
In addition, the service providing method based on the search engine of the embodiment of the invention also has the following additional technical characteristics:
in one embodiment of the invention, the method further comprises:
marking a search request containing service requirements in a search showing log of a search engine;
training the word vector of the search request containing the service requirement by using a preset tool;
and training a logistic regression neural network model as the classifier by taking the word vector as an input layer.
In an embodiment of the present invention, the determining whether a search request input to a search engine by a user includes a service requirement according to a pre-trained classifier includes:
obtaining a word vector of the search request by using a preset tool;
taking the word vector as an input layer of the logistic regression neural network model;
and judging the classification label result through the probability vector of the output layer, and determining whether the search request contains the service requirement.
In an embodiment of the present invention, the obtaining the search feature vector of the search request includes:
obtaining multi-dimensional search features of the search request;
converting the multi-dimensional search features into multi-dimensional vectors by a preset word vector technology;
and splicing the multi-dimensional vectors to obtain the search feature vector.
In one embodiment of the invention, the multi-dimensional search feature comprises at least one of:
word characteristics of the search request, search context Session information, search presentation result information, user click behavior, and user portrait information.
In an embodiment of the present invention, the obtaining a service vector corresponding to the search feature vector according to pre-trained service resource tagging data includes:
matching the pre-trained service resource marking data with the search feature vector;
obtaining a service vector corresponding to the search feature vector according to a matching result, wherein the service vector comprises: and K elements with preset number and sorted according to the matching probability.
In order to achieve the above object, a second embodiment of the present invention provides a service providing device based on a search engine, including:
the judging module is used for judging whether a search request input to a search engine by a user contains a service requirement according to a pre-trained classifier;
the first acquisition module is used for acquiring a search characteristic vector of the search request when the search request contains a service requirement;
the second acquisition module is used for acquiring a service vector corresponding to the search feature vector according to pre-trained service resource marking data;
and the providing module is used for providing the connection service information corresponding to the search request to the user through the search engine according to the service vector.
The service providing device based on the search engine judges whether a search request input to the search engine by a user contains a service requirement according to a pre-trained classifier, if the search request contains the service requirement, a search characteristic vector of the search request is obtained, a service vector corresponding to the search characteristic vector is obtained according to pre-trained service resource marking data, and then connection service information corresponding to the search request is provided to the user through the search engine according to the service vector. Therefore, the service requirements of the users are mined by deeply analyzing the search requests of the users, and the related connection service information is provided for the users in the search engine, so that the function of the search engine is strengthened, and the search results can better meet the requirements of the users.
In addition, the service providing device based on the search engine of the embodiment of the invention also has the following additional technical characteristics:
in one embodiment of the invention, the apparatus further comprises:
the marking module is used for marking a search request containing the service requirement in a search showing log of a search engine;
the first training module is used for training the word vector of the search request containing the service requirement by using a preset tool;
and the second training module is used for training a logistic regression neural network model as the classifier by taking the word vector as an input layer.
In an embodiment of the present invention, the determining module is configured to:
obtaining a word vector of the search request by using a preset tool;
taking the word vector as an input layer of the logistic regression neural network model;
and judging the classification label result through the probability vector of the output layer, and determining whether the search request contains the service requirement.
In an embodiment of the present invention, the first obtaining module is configured to:
obtaining multi-dimensional search features of the search request;
converting the multi-dimensional search features into multi-dimensional vectors by a preset word vector technology;
and splicing the multi-dimensional vectors to obtain the search feature vector.
In one embodiment of the invention, the multi-dimensional search feature comprises at least one of:
word characteristics of the search request, search context Session information, search presentation result information, user click behavior, and user portrait information.
In one embodiment of the invention, the providing module is configured to:
matching the pre-trained service resource marking data with the search feature vector;
obtaining a service vector corresponding to the search feature vector according to a matching result, wherein the service vector comprises: and K elements with preset number and sorted according to the matching probability.
Additional aspects and advantages of the invention will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of the invention.
Drawings
The foregoing and/or additional aspects and advantages of the present invention will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings of which:
FIG. 1 is a flowchart of a search engine based service provision method according to one embodiment of the present invention;
FIG. 2 is a flow diagram of a method for providing search engine based services in accordance with a specific embodiment of the present invention;
FIG. 3 is a schematic diagram of a training process for training a logistic regression neural network, according to one embodiment of the present invention;
FIG. 4 is a schematic structural diagram of a search engine-based service providing apparatus according to an embodiment of the present invention; and
fig. 5 is a schematic structural diagram of a search engine-based service providing apparatus according to an embodiment of the present invention.
Detailed Description
Reference will now be made in detail to embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to the same or similar elements or elements having the same or similar function throughout. The embodiments described below with reference to the drawings are illustrative and intended to be illustrative of the invention and are not to be construed as limiting the invention.
A search engine-based service providing method and apparatus according to an embodiment of the present invention will be described with reference to the accompanying drawings.
Fig. 1 is a flowchart of a search engine-based service providing method according to an embodiment of the present invention. As shown in fig. 1, the method includes:
and S110, judging whether the search request input to the search engine by the user contains service requirements or not according to the pre-trained classifier.
Generally, when a user searches by using a search engine, the user no longer needs to acquire only information, and in many application scenarios, the user also needs to acquire certain service information, for example, when a search request input by the user at the search engine is "taijiong", the user may not only want to acquire information such as a consultation of a movie related to "taijiong", but also may want to acquire service information such as a movie theater in which period in the vicinity of the user, and the movie theater can be played.
Therefore, the service providing method based on the engine search of the embodiment of the invention excavates the service requirement of the user according to the search request of the user, so as to provide the service information for the user according to the service requirement of the user, and better meet the search request of the user.
Specifically, whether the search request input to the search engine by the user contains the service requirement is judged according to a pre-trained classifier, wherein the pre-trained classifier is formed by analyzing and training according to a large amount of search histories and the like, and whether the search request input to the search engine by the user contains the service requirement can be accurately judged.
And S120, if the search request contains the service requirement, acquiring the search feature vector of the search request.
Specifically, if the service requirement is included in the search request, a search feature vector related to the service requirement of the user needs to be acquired, so as to further analyze the service requirement of the user according to the feature vector.
In one embodiment of the invention, if the search request contains a service requirement, multi-dimensional search features of the search request are obtained, wherein the multi-dimensional features of the search request can comprise one or more of word features of the search request, search context Session information, search presentation result information (title, webpage address, webpage abstract and the like in the search result), user click behavior and user portrait information (gender, age, geographic position, search time and the like of the user).
And then converting the multidimensional search features into multidimensional vectors by using a word embedding technology, for example, and then splicing the multidimensional vectors to obtain the search feature vectors.
And S130, acquiring a service vector corresponding to the search feature vector according to the service resource marking data trained in advance.
It can be understood that the pre-trained service resource labeling data stores the search feature vector, the service vector and the corresponding relationship thereof, so that after the search feature vector is obtained, the service vector corresponding to the search feature vector can be obtained according to the pre-trained service resource labeling data.
Specifically, in an embodiment of the present invention, pre-trained service resource tagging data may be matched with a search feature vector, and a service vector corresponding to the search feature vector is obtained according to a matching result, where the service vector includes: and K elements with preset number and sorted according to the matching probability, wherein the elements comprise service related information such as service providers, service subclasses and the like connected with the search request.
It should be noted that the preset number of the service vectors is calibrated according to the size of the font set by the terminal device where the search engine is located, the size of the screen, the interface display mode of the search engine, and the like.
To describe more clearly the specific result of obtaining the service vector corresponding to the search feature vector according to the pre-trained service resource tagging data, the following is exemplified with reference to table 1:
as shown in table 1, when the preset number of the service vectors sorted according to the matching probability is 1 element, and the elements include service providers and service subcategories, and when the search request of the user is a child height increasing recipe, the service vector corresponding to the search feature vector corresponding to the search request is obtained as a soft and rich preferred and child milk powder according to the pre-trained service resource marking data;
when the search request of a user is how long the lung cancer can live in the advanced stage, acquiring a service vector corresponding to a search feature vector corresponding to the search request according to pre-trained service resource marking data, wherein the service vector is good doctor online and corresponds to the lung cancer;
when the search request of the user is a home delivery order query, acquiring service vectors corresponding to the search feature vector corresponding to the search request as an express home and an express home according to pre-trained service resource marking data;
when the search request of the user is complementary food for the baby in eight months, acquiring a service vector corresponding to the search characteristic vector corresponding to the search request as the smooth and rich preferred complementary food according to pre-trained service resource marking data;
when the search request of the user is that the user learns to draw, the service vector corresponding to the search feature vector corresponding to the search request is acquired according to pre-trained service resource marking data, and the user learns and draws with the service vector.
Table 1 example of search requests and service vectors corresponding thereto
| Serial number | Query | Service provider | Service subclass |
| 1 | Recipe for children to increase height | Shunfeng is preferred | Milk powder for children |
| 2 | How long the lung cancer can survive in advanced stage | Good doctor is online | Lung cancer |
| 3 | House emergency delivery single number inquiry | Express home | Express home |
| 4 | Complementary food for eight-month babies | Shunfeng is preferred | Supplementary food |
| 5 | I learn drawing | Who follows to learn | Drawing |
And S140, providing the connection service information corresponding to the search request to the user through the search engine according to the service vector.
Specifically, after the service vector is obtained, connection service information corresponding to the search request is provided for the user through the search engine, so that the third-party service application is not required to be downloaded, the third-party service can be accessed on a platform of the search engine by clicking the relevant connection service information, and the service requirements of the user are met.
In summary, in the service providing method based on the search engine according to the embodiment of the present invention, it is determined whether the search request input to the search engine by the user includes the service requirement according to the pre-trained classifier, if the search request includes the service requirement, the search feature vector of the search request is obtained, the service vector corresponding to the search feature vector is obtained according to the pre-trained service resource tagging data, and then the connection service information corresponding to the search request is provided to the user through the search engine according to the service vector. Therefore, the service requirements of the users are mined by deeply analyzing the search requests of the users, and the related connection service information is provided for the users in the search engine, so that the function of the search engine is strengthened, and the search results can better meet the requirements of the users.
Based on the above embodiments, in order to describe the service providing method based on the search engine more clearly, the service providing method based on the search engine according to the embodiments of the present invention is described in detail below with reference to fig. 2 and a specific training process for the classifier, and the description is as follows:
fig. 2 is a flowchart of a search engine-based service providing method according to an embodiment of the present invention, as shown in fig. 2, the method including:
s210, marking a search request containing service requirements in a search showing log of a search engine.
S220, training a word vector of the search request containing the service requirement by using a preset tool.
Specifically, a search request including a service requirement in a search presentation log is used as a corpus, a word vector including the service requirement is trained, that is, the search request including the service requirement is marked in the search log of a search engine, and then a preset tool, for example, a word2vec tool is used to train the word vector.
And S230, training a logistic regression neural network model as a classifier by taking the word vector as an input layer.
Specifically, a logistic regression neural network model with a three-layer structure is preset and used for distinguishing whether the search request contains the service requirement, wherein after the word vector is obtained, the word vector is used as an input layer to train the edge value of the logistic regression neural network model and the like so as to be capable of distinguishing the search request containing the service requirement from the search request accurately.
For example, as shown in fig. 3, when training the logistic regression neural network model, a large number of word vectors are input into the logistic regression neural network as an input layer, and further, the word vectors are classified into two types for output at an output layer according to the service requirement. Therefore, by inputting a large number of word vectors to train the logistic regression neural network, parameters such as edge values of the logistic regression neural network are continuously improved, and search requests with service requirements and search requests without service requirements can be accurately distinguished by training the logistic regression neural network model.
And S240, acquiring a word vector of the search request by using a preset tool.
And S250, taking the word vector as an input layer of the logistic regression neural network model.
S260, judging the classification label result through the probability vector of the output layer, and determining whether the search request contains the service requirement.
For example, as shown in fig. 3, the result of the classification tag is determined according to the probability vector of the output layer, and it is determined whether the search request includes a service request according to the result of the classification tag.
S270, if the search request contains the service requirement, the search feature vector of the search request is obtained.
Specifically, if the service requirement is included in the search request, a search feature vector related to the service requirement of the user needs to be acquired, so as to further analyze the service requirement of the user according to the feature vector.
And S280, acquiring a service vector corresponding to the search feature vector according to pre-trained service resource marking data.
Specifically, in an embodiment of the present invention, pre-trained service resource tagging data may be matched with a search feature vector, and a service vector corresponding to the search feature vector is obtained according to a matching result, where the service vector includes: and K elements with preset number and sorted according to the matching probability, wherein the elements comprise service related information such as service providers, service subclasses and the like connected with the search request.
And S290, providing the connection service information corresponding to the search request to the user through the search engine according to the service vector.
Specifically, after the service vector is obtained, connection service information corresponding to the search request is provided for the user through the search engine, so that the third-party service application is not required to be downloaded, the third-party service can be accessed on a platform of the search engine by clicking the relevant connection service information, and the service requirements of the user are met.
For example, the user is provided with connection service information of the mei-qu takeout corresponding to the search request "take lunch" through the search engine according to the service vector, so that the user can enter the mei-qu network through the connection service information to order food.
In summary, in the service providing method based on the search engine according to the embodiment of the present invention, the search presentation log based on the search engine is used as the corpus training logistic regression neural network model and is used as the classifier, so as to accurately judge whether the search request input by the user to the search engine includes the service requirement according to the classifier, mine the service requirement of the user, provide the relevant service requirement information for the user, strengthen the function of the search engine, and enable the search result to better satisfy the user requirement.
In order to implement the foregoing embodiments, the present invention further provides a service providing apparatus based on a search engine, and fig. 4 is a schematic structural diagram of the service providing apparatus based on a search engine according to an embodiment of the present invention, as shown in fig. 4, the service providing apparatus based on a search engine includes: a determination module 410, a first acquisition module 420, a second acquisition module 430, and a provision module 440.
The determining module 410 is configured to determine whether a search request input to a search engine by a user includes a service requirement according to a pre-trained classifier.
Specifically, the determining module 410 determines whether the search request input by the user to the search engine includes a service requirement according to a pre-trained classifier, wherein the pre-trained classifier is formed by analyzing and training according to a large amount of search histories and the like, and can accurately determine whether the search request input by the user to the search engine includes the service requirement.
The first obtaining module 420 is configured to obtain a search feature vector of the search request when the search request includes a service requirement.
Specifically, if the search request includes a service requirement, the first obtaining module 420 needs to obtain a search feature vector related to the service requirement of the user, so as to further analyze the service requirement of the user according to the feature vector.
In one embodiment of the present invention, if the search request includes a service requirement, the first obtaining module 420 obtains a multidimensional search characteristic of the search request, wherein the multidimensional search characteristic of the search request may include one or more of a word characteristic of the search request, search context Session information, search presentation result information (a title, a web page address, a web page abstract and the like in the search result), a user click behavior, and user portrait information (a gender, an age, a geographic location, a search time and the like of the user).
And then converting the multidimensional search features into multidimensional vectors by using a word embedding technology, for example, and then splicing the multidimensional vectors to obtain the search feature vectors.
The second obtaining module 430 is configured to obtain a service vector corresponding to the search feature vector according to pre-trained service resource tagging data.
Specifically, in an embodiment of the present invention, the second obtaining module 430 may match pre-trained service resource tagging data with the search feature vector, and obtain a service vector corresponding to the search feature vector according to a matching result, where the service vector includes: and K elements with preset number and sorted according to the matching probability, wherein the elements comprise service related information such as service providers, service subclasses and the like connected with the search request.
And a providing module 440, configured to provide the connection service information corresponding to the search request to the user through the search engine according to the service vector.
Specifically, after the service vector is obtained, the providing module 440 provides the connection service information corresponding to the search request to the user through the search engine, so that the third-party service application is not required to be downloaded, and the third-party service can be accessed on the platform of the search engine by clicking the relevant connection service information, thereby meeting the service requirement of the user.
In summary, the service providing apparatus based on a search engine according to the embodiment of the present invention determines whether a search request input to the search engine by a user includes a service requirement according to a pre-trained classifier, obtains a search feature vector of the search request if the search request includes the service requirement, obtains a service vector corresponding to the search feature vector according to pre-trained service resource tagging data, and provides connection service information corresponding to the search request to the user through the search engine according to the service vector. Therefore, the service requirements of the users are mined by deeply analyzing the search requests of the users, and the related connection service information is provided for the users in the search engine, so that the function of the search engine is strengthened, and the search results can better meet the requirements of the users.
Based on the above embodiments, in order to describe the service providing apparatus based on the search engine more clearly, the following describes the service providing apparatus based on the search engine in detail by referring to fig. 5 and a specific training process for the classifier, and the following description is made:
fig. 5 is a schematic structural diagram of a search engine-based service providing apparatus according to an embodiment of the present invention, and as shown in fig. 5, the search engine-based service providing apparatus includes, based on fig. 4: an annotation module 450, a first training module 460, and a second training module 470.
The annotation module 450 is configured to annotate a search request containing a service requirement in a search presentation log of a search engine.
The first training module 460 is used to train a word vector containing a search request for service requirements using preset tools.
Specifically, taking the search request containing the service requirement in the search presentation log as a corpus, the first training module 460 trains a word vector containing the service request, i.e., marks the search request containing the service requirement in the search log of the search engine, and further trains the word vector using a preset tool, for example, using a word2vec tool.
The second training module 470 is used to train the logistic regression neural network model as a classifier with the word vectors as an input layer.
Specifically, a logistic regression neural network model with a three-layer structure is preset, and the logistic regression neural network model is used for distinguishing whether the search request includes the service requirement, wherein, in order to enable the logistic regression neural network model to accurately distinguish whether the search request includes the service requirement, after the word vector is obtained, the second training module 470 trains the edge value and the like of the logistic regression neural network model by taking the word vector as an input layer, so that the trained logistic regression neural network model can be used as a classifier for distinguishing the search request with the service requirement from the search request without the service requirement.
Further, the determining module 410 obtains a word vector of the search request by using a preset tool, takes the word vector as an input layer of the logistic regression neural network model, determines the classification label result according to the probability vector of the output layer, and determines whether the search request includes the service requirement.
Thus, if the service requirement is included in the search request, the first obtaining module 420 needs to obtain a search feature vector related to the service requirement of the user, so as to further analyze the service requirement of the user according to the feature vector.
Further, after the first obtaining module 420 obtains the search feature vector related to the service requirement of the user, the providing module 440 may match the pre-trained service resource tagging data with the search feature vector, and obtain the service vector corresponding to the search feature vector according to the matching result, so that the providing module 440 provides the connection service information corresponding to the search request to the user through the search engine according to the service vector.
Wherein the service vector comprises: and K elements with preset number and sorted according to the matching probability, wherein the elements comprise service related information such as service providers, service subclasses and the like connected with the search request.
It should be noted that, the service providing apparatus based on a search engine according to the embodiment of the present invention corresponds to the service providing method based on a search engine described above with reference to fig. 1 to 3, and details that are not disclosed in the service providing apparatus based on a search engine according to the embodiment of the present invention are not repeated herein.
In summary, in the service providing apparatus based on a search engine according to the embodiment of the present invention, the search presentation log based on the search engine is used as the corpus training logistic regression neural network model and is used as the classifier, so as to accurately judge whether the search request input by the user to the search engine includes the service requirement according to the classifier, mine the service requirement of the user, provide the relevant service requirement information for the user, and enhance the function of the search engine, so that the search result more satisfies the user requirement.
Furthermore, the terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one such feature. In the description of the present invention, "a plurality" means at least two, e.g., two, three, etc., unless specifically limited otherwise.
In the description herein, references to the description of the term "one embodiment," "some embodiments," "an example," "a specific example," or "some examples," etc., mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above are not necessarily intended to refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Furthermore, various embodiments or examples and features of different embodiments or examples described in this specification can be combined and combined by one skilled in the art without contradiction.
Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention, and that variations, modifications, substitutions and alterations can be made to the above embodiments by those of ordinary skill in the art within the scope of the present invention.
Claims (6)
1. A service providing method based on a search engine is characterized by comprising the following steps:
judging whether a search request input to a search engine by a user contains a service requirement according to a pre-trained classifier, wherein the search request does not contain the name of a service provider corresponding to the service requirement, and the service requirement is an extended service request corresponding to the search request;
if the search request contains service requirements, obtaining multi-dimensional search features of the search request;
converting the multi-dimensional search features into multi-dimensional vectors by a preset word vector technology;
splicing the multi-dimensional vectors to obtain a search feature vector of the search request, wherein the multi-dimensional search feature comprises at least one of the following: word characteristics of the search request, search context Session information, search presentation result information, user click behavior, and user portrait information;
matching pre-trained service resource marking data with the search feature vector;
obtaining a service vector corresponding to the search feature vector according to a matching result, wherein the service vector comprises: k elements with preset number are sorted according to the matching probability, and the elements comprise service information such as service providers, service subclasses and the like connected with the search request;
and providing connection service information corresponding to the search request to the user through the search engine according to the service vector.
2. The method of claim 1, further comprising:
marking a search request containing service requirements in a search showing log of a search engine;
training the word vector of the search request containing the service requirement by using a preset tool;
and training a logistic regression neural network model as the classifier by taking the word vector as an input layer.
3. The method of claim 2, wherein determining whether a search request input by a user to a search engine contains a service requirement based on a pre-trained classifier comprises:
acquiring a word vector of a search request input to a search engine by the user by using a preset tool;
taking a word vector of a search request input to a search engine by the user as an input layer of the logistic regression neural network model;
and judging the classification label result through the probability vector of the output layer, and determining whether the search request contains the service requirement.
4. A service providing apparatus based on a search engine, comprising:
the system comprises a judging module, a searching module and a searching module, wherein the judging module is used for judging whether a search request input to a search engine by a user contains a service requirement according to a pre-trained classifier, the search request does not contain the name of a service provider corresponding to the service requirement, and the service requirement is an extended service request corresponding to the search request;
the first obtaining module is configured to obtain multidimensional search features of the search request when the search request includes a service requirement, convert the multidimensional search features into multidimensional vectors by using a preset word vector technology, and splice the multidimensional vectors to obtain the search feature vectors of the search request, where the multidimensional search features include at least one of the following: word characteristics of the search request, search context Session information, search presentation result information, user click behavior, and user portrait information;
a second obtaining module, configured to match pre-trained service resource tagging data with the search feature vector, and obtain a service vector corresponding to the search feature vector according to a matching result, where the service vector includes: k elements with preset number are sorted according to the matching probability, and the elements comprise service information such as service providers, service subclasses and the like connected with the search request;
and the providing module is used for providing the connection service information corresponding to the search request to the user through the search engine according to the service vector.
5. The apparatus of claim 4, further comprising:
the marking module is used for marking a search request containing the service requirement in a search showing log of a search engine;
the first training module is used for training the word vector of the search request containing the service requirement by using a preset tool;
and the second training module is used for training a logistic regression neural network model as the classifier by taking the word vector as an input layer.
6. The apparatus of claim 5, wherein the determination module is to:
acquiring a word vector of a search request input to a search engine by the user by using a preset tool;
taking a word vector of a search request input to a search engine by the user as an input layer of the logistic regression neural network model;
and judging the classification label result through the probability vector of the output layer, and determining whether the search request contains the service requirement.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610630338.4A CN106326338B (en) | 2016-08-03 | 2016-08-03 | Service providing method and device based on search engine |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610630338.4A CN106326338B (en) | 2016-08-03 | 2016-08-03 | Service providing method and device based on search engine |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106326338A CN106326338A (en) | 2017-01-11 |
| CN106326338B true CN106326338B (en) | 2020-02-04 |
Family
ID=57739582
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610630338.4A Active CN106326338B (en) | 2016-08-03 | 2016-08-03 | Service providing method and device based on search engine |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106326338B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10747825B2 (en) * | 2017-02-27 | 2020-08-18 | Google Llc | Content search engine |
| CN110309114B (en) * | 2018-02-28 | 2021-07-27 | 腾讯科技（深圳）有限公司 | Method and device for processing media information, storage medium and electronic device |
| CN110134798B (en) * | 2019-05-15 | 2021-07-27 | 八戒科技服务有限公司 | Innovation demand mining method, system, terminal and medium |
| CN114223011A (en) * | 2019-10-23 | 2022-03-22 | 深圳市欢太科技有限公司 | Entry pushing method and device, electronic equipment and storage medium |
| CN111310008A (en) * | 2020-03-20 | 2020-06-19 | 北京三快在线科技有限公司 | Search intention recognition method and device, electronic equipment and storage medium |
| CN111324819B (en) * | 2020-03-24 | 2021-07-30 | 北京字节跳动网络技术有限公司 | Method and device for searching media content, computer equipment and storage medium |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1437142A (en) * | 2002-02-03 | 2003-08-20 | 张夫杰 | Medicine search method and system in Internet |
| CN102426572A (en) * | 2011-07-05 | 2012-04-25 | 百度在线网络技术（北京）有限公司 | Method and equipment for classifying business entries |
| CN102737050A (en) * | 2011-04-11 | 2012-10-17 | 阿里巴巴集团控股有限公司 | Keyword dynamic regulating method and system applied in search engine optimization |
| CN103336796A (en) * | 2013-06-09 | 2013-10-02 | 北京百度网讯科技有限公司 | Method and system for displaying door buster directly |
| CN103729365A (en) * | 2012-10-12 | 2014-04-16 | 阿里巴巴集团控股有限公司 | Searching method and system |
| CN104615767A (en) * | 2015-02-15 | 2015-05-13 | 百度在线网络技术（北京）有限公司 | Searching-ranking model training method and device and search processing method |
| CN104794246A (en) * | 2015-05-15 | 2015-07-22 | 百度在线网络技术（北京）有限公司 | Information search method and information search device |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160189026A1 (en) * | 2014-12-26 | 2016-06-30 | Universidad De Santiago De Chile | Running Time Prediction Algorithm for WAND Queries |
-
2016
- 2016-08-03 CN CN201610630338.4A patent/CN106326338B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1437142A (en) * | 2002-02-03 | 2003-08-20 | 张夫杰 | Medicine search method and system in Internet |
| CN102737050A (en) * | 2011-04-11 | 2012-10-17 | 阿里巴巴集团控股有限公司 | Keyword dynamic regulating method and system applied in search engine optimization |
| CN102426572A (en) * | 2011-07-05 | 2012-04-25 | 百度在线网络技术（北京）有限公司 | Method and equipment for classifying business entries |
| CN103729365A (en) * | 2012-10-12 | 2014-04-16 | 阿里巴巴集团控股有限公司 | Searching method and system |
| CN103336796A (en) * | 2013-06-09 | 2013-10-02 | 北京百度网讯科技有限公司 | Method and system for displaying door buster directly |
| CN104615767A (en) * | 2015-02-15 | 2015-05-13 | 百度在线网络技术（北京）有限公司 | Searching-ranking model training method and device and search processing method |
| CN104794246A (en) * | 2015-05-15 | 2015-07-22 | 百度在线网络技术（北京）有限公司 | Information search method and information search device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106326338A (en) | 2017-01-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106326338B (en) | Service providing method and device based on search engine | |
| CN108009228B (en) | Method and device for setting content label and storage medium | |
| US10346484B2 (en) | Combining website characteristics in an automatically generated website | |
| CN110134931B (en) | Medium title generation method, medium title generation device, electronic equipment and readable medium | |
| US9727584B2 (en) | Refining image annotations | |
| CN110597962B (en) | Search result display method and device, medium and electronic equipment | |
| WO2016179938A1 (en) | Method and device for question recommendation | |
| US20170228459A1 (en) | Method and device for mobile searching based on artificial intelligence | |
| WO2015185019A1 (en) | Semantic comprehension-based expression input method and apparatus | |
| US9720904B2 (en) | Generating training data for disambiguation | |
| US8873867B1 (en) | Assigning labels to images | |
| CN107562939B (en) | Vertical domain news recommendation method and device and readable storage medium | |
| US20140095308A1 (en) | Advertisement distribution apparatus and advertisement distribution method | |
| CN103577534B (en) | Searching method and search engine | |
| CN103577549A (en) | Crowd portrayal system and method based on microblog label | |
| CN103853757B (en) | The information displaying method and system of network, terminal and information show processing unit | |
| CN104866554B (en) | A kind of individuation search method and system based on socialization mark | |
| US20150006570A1 (en) | Search apparatus, search method, search program, and recording medium | |
| US20130232170A1 (en) | Related-word registration device, information processing device, related-word registration method, program for related-word registration device, recording medium, and related-word registration system | |
| US11574123B2 (en) | Content analysis utilizing general knowledge base | |
| CN105653547A (en) | Method and device for extracting keywords of text | |
| CN106815253B (en) | Mining method based on mixed data type data | |
| CN104881428A (en) | Information graph extracting and retrieving method and device for information graph webpages | |
| CN109582868A (en) | The search recommended method of preference is clicked based on term vector weighting, support vector regression and user | |
| Arabshian et al. | Lexont: A semi-automatic ontology creation tool for programmable web |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |