CN111143582A - Multimedia resource recommendation method and device for updating associative words in real time through double indexes - Google Patents
Multimedia resource recommendation method and device for updating associative words in real time through double indexes Download PDFInfo
- Publication number
- CN111143582A CN111143582A CN201911228998.XA CN201911228998A CN111143582A CN 111143582 A CN111143582 A CN 111143582A CN 201911228998 A CN201911228998 A CN 201911228998A CN 111143582 A CN111143582 A CN 111143582A
- Authority
- CN
- China
- Prior art keywords
- information
- index
- word
- multimedia resource
- association
- 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.)
- Granted
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
- G06F16/40—Information retrieval; Database structures therefor; File system structures therefor of multimedia data, e.g. slideshows comprising image and additional audio data
- G06F16/43—Querying
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/40—Information retrieval; Database structures therefor; File system structures therefor of multimedia data, e.g. slideshows comprising image and additional audio data
- G06F16/41—Indexing; Data structures therefor; Storage structures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/40—Information retrieval; Database structures therefor; File system structures therefor of multimedia data, e.g. slideshows comprising image and additional audio data
- G06F16/48—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
Abstract
The invention relates to the technical field of Internet, in particular to a multimedia resource recommendation method and a multimedia resource recommendation device for updating association words in real time by double indexes, which are used for solving the problem that the index information of the association words cannot be updated correctly due to the fact that the change condition of media resources cannot be counted in real time, and the method comprises the following steps: responding to an association word query request sent by intelligent equipment, searching in a global index library, wherein the global index library is established based on index information in a real-time index library updated in real time, acquiring the association word index information corresponding to the association word query request based on a search result, further acquiring corresponding multimedia resources, and then sending the multimedia resources to the intelligent equipment. Therefore, the consistency between the actual attribute information of the multimedia resources and the attribute information recorded in the index library is ensured based on the global index library updated in real time, the accuracy of the multimedia resources obtained by the intelligent equipment is also ensured, and the processing efficiency is improved.
Description
Technical Field
The application relates to the technical field of internet, in particular to a multimedia resource recommendation method and device for updating associative words in real time through double indexes.
Background
With the development of internet technology, people can browse multimedia resources, hereinafter referred to as media resources, published on different media platforms on an intelligent television, users can quickly match corresponding association words based on the input of the users by inputting pinyin information or partial character information on a search interface of the intelligent television, and further, the acquirable media resources are searched in an index library based on the association words selected by the users and are recommended to the users.
In the prior art, an index library generally establishes index information based on association words and attribute information of associated media resources, the association words are obtained by extracting and calculating texts such as title tags of each media resource according to a natural language processing principle, the attribute information specifically includes attribute information of each media resource from which the association words can be extracted, the attribute information includes language information corresponding to the media resource, media platform information to which the association words belong, service type information to which the association words belong, and terminal equipment type and model information of a smart television that can normally display the media resources, and each index information in the index library, that is, the association words and the attribute information of all associated media resources, is updated with "day" as an update period.
However, since the update of the media resource occurs in real time, the attribute information of the media resource and the corresponding information such as title and tag may change, so that, based on the index library which is not updated in time, when the attribute information of the media resource changes, the media resource corresponding to the association word cannot be obtained, and the search abnormality is displayed, on the other hand, by using the update method provided in the prior art, when the attribute information of one media resource associated with the association word changes, the update is completed, which may cause the change of the attribute information of other media resources associated with the association word, the media resource which can be normally viewed cannot be searched in the index library based on the association word selected by the user, and since the update of the index information in the index library is implemented by calculating based on all the media resources during the update, the calculation efficiency is low, and the method can not adapt to the real-time updating status of the media resources, can not ensure the consistency of the media resource attribute information associated with the association words in the index library and the media resource attribute information after real-time updating, and greatly influences the user experience.
In view of the above, a new method for recommending multimedia resources is needed to solve the above problems.
Disclosure of Invention
The embodiment of the invention provides a multimedia resource recommendation method and device for updating association words in real time through double indexes, which are used for solving the problem that the index information of the association words cannot be updated correctly due to the fact that the change condition of media resources cannot be counted in real time in the prior art.
The embodiment of the invention provides the following specific technical scheme:
a multimedia resource recommendation method for updating associative words in real time through double indexes comprises the following steps:
responding to an association word query request sent by intelligent equipment, searching in a global index library, wherein the association word query request at least comprises an equipment model of the intelligent equipment and user identity identification information, the global index library is established based on index information in a real-time index library, the index information in the real-time index library is updated in real time based on the change of multimedia resource information in a network, and one piece of index information comprises at least one association word extracted from one multimedia resource and attribute information of the one multimedia resource;
acquiring association word index information corresponding to the association word query request based on a search result, and acquiring corresponding multimedia resources based on the association word index information;
and sending the multimedia resource to the intelligent equipment.
A multimedia resource recommendation device for updating associative words in real time through double indexes comprises the following components:
the searching unit is used for responding to an association word query request sent by intelligent equipment and searching in a global index library, wherein the association word query request at least comprises the equipment model of the intelligent equipment and user identity identification information, the global index library is established based on index information in a real-time index library, the index information in the real-time index library is updated in real time based on the change of multimedia resource information in a network, and one piece of index information comprises at least one association word extracted from one multimedia resource and attribute information of the one multimedia resource;
the acquisition unit is used for acquiring association word index information corresponding to the association word query request based on a search result and acquiring corresponding multimedia resources based on the association word index information;
and the sending unit is used for sending the multimedia resource to the intelligent equipment.
The invention has the following beneficial effects:
in the embodiment of the application, in response to an association word query request sent by intelligent equipment, searching is performed in a global index library, the global index library is established based on index information in a real-time index library, the index information in the real-time index library is updated in real time based on the change of multimedia resource information in a network, then, based on a search result, association word index information corresponding to the association word query request is obtained, corresponding multimedia resources are obtained based on the association word index information, and then, the multimedia resources are sent to the intelligent equipment. Therefore, the consistency between the actual attribute information of the multimedia resources and the attribute information recorded in the index library is ensured based on the global index library updated in real time, the accuracy of the multimedia resources obtained by the intelligent equipment is also ensured, the processing efficiency is improved, and the user experience is improved.
Drawings
FIG. 1 is a schematic flow chart illustrating the real-time index library establishment in the embodiment of the present application;
FIG. 2 is a schematic flow chart illustrating establishment of a global index repository in an embodiment of the present application;
FIG. 3 is a schematic flowchart illustrating a process of recommending multimedia resources to an intelligent device according to an embodiment of the present application;
FIG. 4 is a schematic diagram of a logical structure of a server in an embodiment of the present application;
fig. 5 is a schematic physical structure diagram of a server in the embodiment of the present application.
Detailed Description
In order to solve the problem that the change condition of media resources cannot be counted in real time in the prior art, so that the association word information cannot be updated correctly, the method and the device respond to an association word query request sent by intelligent equipment, search is carried out in a global index library, the association word query request at least comprises the equipment model of the intelligent equipment and user identity identification information, the global index library is established based on index information in a real-time index library, the index information in the real-time index library is updated in real time based on the change of multimedia resource information in a network, and one index information corresponds to one association word and attribute information of multimedia resources related to the association word; acquiring association word index information corresponding to the association word query request based on a search result, and acquiring corresponding multimedia resources based on the association word index information; and finally, sending the multimedia resource to the intelligent equipment.
Preferred embodiments of the present application will now be described with reference to the accompanying drawings:
to ensure the completeness of the scheme, the process of establishing the real-time index library is described as follows:
the server establishes a real-time index library based on the attribute information of the multimedia resources and the association words extracted from the titles or labels of the multimedia resources.
In some embodiments, a server processes an acquired multimedia resource which changes in real time by using a wave-tide (Storm) streaming cluster real-time computing manner, and calculates an information Digest key of the multimedia resource by using an information Digest Algorithm (MD 5) Algorithm based on attribute information of the multimedia resource, where the information Digest key is used to represent key attribute information of the multimedia resource, and at the same time, defines a User defined processing function (UDF) based on a stripe-by-stripe extraction logic of data by a data warehouse tool (HIVE) during offline processing, and then extracts at least one associative word from text information such as a title, a tag, and the like of the multimedia resource based on the UDF. And then generating index information based on the at least one association word, the attribute information of the multimedia resource, the key attribute information of the multimedia resource and the information abstract key of the multimedia resource, sequentially generating corresponding amounts of index information based on the acquired attribute information of all the multimedia resources, and sequentially storing the index information into a real-time index library which is preset and allocated with a storage space.
Specifically, referring to fig. 1, taking a multimedia resource as an example, the server reads a multimedia resource that changes in the network in real time, and performs the following operations on each read multimedia resource:
step 101: extracting at least one association word contained in one multimedia resource, and acquiring ID information contained in attribute information of the one multimedia resource.
In some embodiments, the server uses a Storm streaming cluster capable of performing real-time computation to process the obtained changed multimedia resources in the network, where the changed multimedia resources may be completely new online multimedia resources or multimedia resources with changed attribute information, and the attribute information includes all feature information capable of representing the multimedia resources.
Firstly, the server extracts UDF defined by logic item by item of HIVE data during off-line processing, and extracts at least one associative word from the text information such as the title or the label of the multimedia resource. Meanwhile, acquiring attribute information of the multimedia resource, wherein the attribute information comprises all characteristic information of the multimedia resource, the attribute information may be further divided into key attribute information and non-key attribute information, where the key attribute information includes information such as title information of a multimedia resource, Identity Information (ID) of the multimedia resource, online identification information of the multimedia resource, terminal model information supported by the multimedia resource, media platform information to which the multimedia resource belongs, and terminal type information supported by the multimedia resource, the non-key attribute information comprises information such as time information, place information and played time length information of the multimedia resource, and the corresponding multimedia resource is not influenced by the change of the non-key attribute information.
Step 102: and obtaining an information digest key based on the attribute information of the multimedia resource.
Specifically, the server generates a 64-bit information digest key by using an information digest generation algorithm MD5 based on the key attribute information of the multimedia resource, and can determine whether the key attribute information of the multimedia resource changes based on the information digest key, where the change in the key attribute information may cause that the corresponding multimedia resource cannot be accurately found.
Step 103: is it determined whether the one multimedia asset is not recorded in the real-time index repository based on the ID information? If so, go to step 104, otherwise, go to step 108.
The server calls a Search engine service (Elastic Search, ES) to Search in a real-time index library based on the ID information of the multimedia resources contained in the obtained attribute information of the multimedia resources, if the index information matched with the ID information of the multimedia resources exists in the current real-time index library, the multimedia resources are judged to be recorded in the real-time index library, otherwise, if the index information matched with the ID information of the multimedia resources does not exist in the current real-time index library, the multimedia resources are judged not to be recorded in the real-time index library.
Step 104: is the multimedia asset determined to be online? If yes, go to step 106; otherwise, step 105 is performed.
Optionally, when the server queries based on the ID information of the multimedia resource, after determining that the multimedia resource is not recorded in the real-time index library, the server may further determine based on online identification information of the multimedia resource, where the online identification information is used to represent whether the multimedia resource is online, and the online identification information belongs to key attribute information of the multimedia resource, and specifically is field information of the multimedia resource. If the result of the representation of the online identification information is that the multimedia resource is online, corresponding index information can be further generated based on the multimedia resource, and if the result of the representation of the online identification information is that the multimedia resource is not online, corresponding index information cannot be generated based on the multimedia resource.
For example, for a multimedia resource X obtained in real time, based on ID information of the multimedia resource X, the ES is called to search in a real-time index library, and it is determined that the multimedia resource X is not recorded in the real-time index library, then further, online identification information of the multimedia resource X is obtained, and if it is determined that the multimedia resource X is not online based on the online identification information, corresponding index information will not be generated based on the certain multimedia resource.
Similarly, the index information may also be directly generated based on a multimedia resource that is not recorded in the real-time index library, and in order to ensure effective utilization of the storage space, optionally, the determination in step 103 may be performed, and when it is determined that the multimedia resource is online, the subsequent operation of generating the index information is performed.
Step 105: index information is not established based on the multimedia resources.
Specifically, when the server determines that the multimedia resource is not online or the key attribute information of the multimedia resource is not changed, the server does not establish the index information based on the multimedia resource.
For example, the server determines that index information corresponding to a multimedia resource a exists in a real-time index library, and learns that play duration information and resource release position information in attribute information of the multimedia resource a are changed, but since the play duration information and the resource release position information are non-key attribute information of the multimedia resource a, the server does not re-process the multimedia resource a to establish the index information.
Step 106: and generating index information based on the at least one association word and the attribute information of the piece of multimedia resource.
Specifically, after determining that the multimedia resource is not recorded in the real-time index library and the multimedia resource is online, the server generates an index message based on at least one association word extracted from the title information of the multimedia resource and the attribute information of the multimedia resource, where one multimedia resource generates one piece of index message correspondingly, and the content of the one piece of index message specifically includes: the multimedia resource management system comprises at least one associative word, attribute information of the multimedia resource, key attribute information of the multimedia resource and an information digest key generated based on the key attribute information, wherein the information digest key is calculated by adopting an MD5 algorithm based on the key attribute information, and the key attribute information is contained in the attribute information.
Step 107: and storing the index information to the real-time index library.
Specifically, the server determines that the multimedia resource is not recorded in the real-time index library based on the ID information and the online identification information included in the attribute information of the multimedia resource, and generates an index information based on the multimedia resource after the multimedia resource is online, and further stores the index information in the real-time index library to which a storage space is pre-allocated.
Step 108: and determining index information corresponding to the multimedia resource in a real-time index library based on the ID information.
Specifically, the server invokes an ES search engine based on the ID information of the multimedia resource included in the obtained attribute information of the multimedia resource, and determines the index information corresponding to the piece of multimedia resource existing in the real-time index library.
Step 109: and acquiring a historical information digest key of the multimedia resource and at least one historical association word included in the index information.
Specifically, the server searches in a real-time index library to obtain index information corresponding to one multimedia resource based on ID information of the one multimedia resource, and then obtains at least one history association word and a history information digest key, which are included in the index information and generated based on attribute information of the one multimedia resource before the change occurs, from the index information.
Step 110: is the information digest key and the history digest key different in value? If yes, go to step 111, otherwise go to step 105.
Specifically, for a multimedia resource which is acquired in real time, has generated corresponding index information and is recorded in a real-time index library, and attribute information of which changes, after determining corresponding index information in the real-time index library based on ID information of the multimedia resource, the server acquires a history information digest key of the multimedia resource contained in the index information. And comparing the historical information digest key with the information digest key obtained by recalculating the key attribute information of the multimedia resources acquired in real time by using the MD5 algorithm, if the values of the historical information digest key and the information digest key are the same, determining that the key attribute information of the multimedia resource is not changed, otherwise, determining that the key attribute information of the multimedia resource is changed if the values of the historical information digest key and the information digest key are different, wherein the change of the key attribute information may be that the extracted association word of the multimedia resource is changed, or, part of the key attribute information of the multimedia resource is changed, or both the associative word and part of the key attribute information that can be extracted by the multimedia resource are changed. The key attribute information comprises information such as title information of the multimedia resource, ID information of the multimedia resource, model information of a terminal supported by the multimedia resource, media platform information to which the multimedia resource belongs, and terminal type information supported by the multimedia resource.
For example, a server acquires a multimedia resource B with changed attribute information in real time by using a storm streaming cluster, acquires the attribute information of the multimedia resource B, acquires information digest key of the multimedia resource B by using an MD5 algorithm based on the key attribute information of the multimedia resource B, acquires ID information of the multimedia resource B by using an ES search engine based on the ID information, finds corresponding index information in a real-time index library, acquires a historical information digest key B1 included in the index information, compares the information digest key B with the historical information digest key B1, and if the values of the information digest key B are different from the values of the information digest key B1, indicates that the key attribute information of the multimedia resource B has changed.
Step 111: is the at least one association word different from the at least one historical association word determined? If yes, go to step 112, otherwise go to step 113.
Specifically, the server extracts at least one association word from a title or a label of a multimedia resource by using a UDF function for the multimedia resource obtained in real time, searches corresponding index information in a real-time index library based on ID information of the multimedia resource, can further obtain at least one historical association word included in the index information, and compares the at least one association word with the at least one historical association word. If the association words are found to be not identical, the association words extracted from the multimedia resource are judged to be changed, otherwise, if the association words are found to be identical to the at least one historical association word, the association words extracted from the multimedia resource are judged not to be changed.
For example, for a multimedia resource C, a UDF algorithm is previously adopted to extract an association word a and an association word b, the generated association words are associated with attribute information of the multimedia resource C and then stored as index information in a real-time index library, and if a changing multimedia resource C is obtained in real time, the association words a and b are historical association words of the multimedia resource C. And aiming at the newly acquired multimedia resource C, extracting an association word b and an association word C by adopting a UDF algorithm, and then, comparing to find that the historical association words are different from the currently extracted association words, indicating that the association words extracted by the multimedia resource are changed.
Step 112: and recording the at least one historical associated word, and updating the index information based on the at least one associated word and the attribute information.
Specifically, after determining that the key attribute information of the multimedia resource changes based on the information abstract, the server determines that the association word extracted from the title of the multimedia resource also changes, and records at least one historical association word of the multimedia resource, and updates a corresponding piece of index information in the real-time index library based on the re-extracted at least one association word and the attribute information of the multimedia resource, where one piece of index information includes the updated at least one association word, the recorded changed at least one historical association word, the updated attribute information of the multimedia resource, the updated key attribute information of the multimedia resource, and the updated information abstract key generated based on the key attribute information.
For example, the real-time index library has index information Y established based on the multimedia resource D, and the associative words extracted based on the title information are: history association words p, history association words q, history attribute information O, history key attribute information O and history information digest key 1; in the process of processing the changed multimedia resource in real time, the multimedia resource D is obtained again, and the corresponding calculation shows that the association words extracted by the multimedia resource D change, and the newly extracted association words are: the association word p, the association word z, and the attribute information change are: the server can determine that the historical association word q cannot be extracted from the multimedia resource D, so that the current historical association word q is recorded, the association word z is added to the index information, and the corresponding index information in the real-time index is updated based on all changed information, that is, the index information Y includes: the method comprises the following steps of associating words p, associating words z, attribute information H, key attribute information H, an information abstract key 2 and recorded history associating words q.
Step 112: updating the index information based on the attribute information of the one multimedia resource.
Specifically, the server determines, based on ID information of a multimedia resource, that a historical information digest key recorded in corresponding index information in a real-time index library is different in value from a newly generated information digest key, but when the at least one association word is the same as the at least one historical association word, updates, based on attribute information of the multimedia resource, index information corresponding to the multimedia resource and existing in the real-time index library, and generates updated index information, where one piece of index information includes the at least one association word, the updated attribute information of the multimedia resource, the updated key attribute information of the multimedia resource, and an information digest key generated based on the updated key attribute information of the multimedia resource.
For example, the real-time index library has index information X established based on the multimedia resource E, and the associative words extracted based on the title information are: an association word d, an association word e, attribute information M, key attribute information M and an information digest key 3; in the process of processing the changed multimedia resource in real time, the multimedia resource E is obtained again, and through corresponding calculation, the association words extracted by the multimedia resource E are not changed, and the attribute information is changed as follows: attribute information N, key attribute information N, and information digest key 4, the server updates the corresponding index information in the real-time index based on only the changed information, that is, the index information X includes the following contents: the method comprises the following steps of associating words d, associating words e, attribute information N, key attribute information N and an information digest key 4.
Therefore, after one multimedia resource is processed, all multimedia resources acquired in real time can be sequentially processed in the same way, so that the index information in the real-time index library is increased or updated.
Based on the process of increasing or updating the index information in the real-time index base in real time and based on the changed index information in the real-time index base, the new creation or updating of the index information of the association words included in the global index base can be further realized, and the real-time performance of the updating is ensured.
The updating of the real-time index base results in updating of the global index base, and the real-time updating process of the global index base can be regarded as being established based on the index information of real-time change in the real-time index base. Each piece of association word index information in the global index library comprises an association word and global information of all multimedia resources capable of extracting the association word, the global attribute information is a part of the multimedia resource attribute information, and corresponding multimedia resources can be accurately found based on the global attribute information.
The following describes a process of updating the global index repository in a real-time update manner by taking a piece of changed index information as an example, with reference to fig. 2:
step 201: and acquiring a piece of index information which changes in the real-time index database in real time.
Specifically, based on the change of the index information in the real-time index library, the server obtains a piece of index information that has changed in the real-time index library, where the reason for the change of the index information includes that the multimedia resource that has correspondingly generated the index information is a newly online multimedia resource that is not recorded in the real-time index library, or a multimedia resource whose key attribute information has changed.
Step 202: and extracting at least one association word contained in the index information and the attribute information of the associated multimedia resource.
Specifically, after acquiring a piece of changed index information, the server extracts all information included in the index information, specifically, at least one association word, attribute information of an associated multimedia resource, key attribute information of the multimedia resource, and an information digest key generated based on the key attribute information.
Step 203: and determining the global attribute information of the multimedia resource based on the attribute information of the multimedia resource.
After extracting the attribute information of the associated multimedia resource contained in the changed index information, the server determines the global attribute information of the multimedia resource based on the attribute information, wherein the global attribute information at least comprises: the language information of the multimedia resource, the media platform information to which the multimedia resource belongs, the terminal type information supported by the multimedia resource, the terminal model information supported by the multimedia resource, the service type information corresponding to the multimedia resource, and the ID information of the multimedia resource.
Step 204: and reading an associated word contained in the index information.
Specifically, after acquiring the changed index information, the server acquires at least one association word included in the index information. Furthermore, an associative word is read first, and subsequent operations are performed based on the associative word.
Step 205: is the determination made as to whether the associated word has matching associated word index information in the global index repository? If so, go to step 206, otherwise, go to step 210.
And the server searches in a global index library by adopting an ES search engine based on the obtained association word, and determines whether association word index information matched with the association word exists in the global index library, wherein the association word index information comprises the association word and global attribute information of the multimedia resource capable of extracting the association word. Specifically, when it is determined that the obtained one associated word is the same as an associated word included in the index information of one associated word in the global index library, it is determined that the one associated word has matched index information of the associated word in the global index library, and otherwise, it is determined that the one associated word does not have matched index information of the associated word in the global index library.
Step 206: is it determined whether the one associative word is recorded as a changed history associative word in the real-time index repository? If yes, go to step 207, otherwise, go to step 208.
Specifically, the association words included in the index information of the real-time index library may include at least one recorded history association word that has changed and at least one re-extracted association word. The related content of the recorded history association words is already described in detail in step 112, and is not described in detail here. The server obtains an association word and further judges whether the association word is recorded as a changed history association word, and because the changed history association word corresponding to the multimedia resource does not have an association relation with the multimedia resource any more, the existence state of the association word in the global index library needs to be processed.
Step 207: and updating corresponding association word index information in the global index library based on the one association word.
Specifically, when the server determines that one currently acquired association word is recorded as a changed history association word, the server updates the association word index information matched in the global index library based on the one association word, and specifically, removes global attribute information of the multimedia resource once associated with the one association word from the association word index information.
Step 208: and associating the association word with the global attribute information to generate an association word index element.
Specifically, when the server determines that a read association word is not recorded as a changed history association word in the real-time index library, the server associates the association word with global attribute information generated based on attribute information of a corresponding multimedia resource to generate an association word index element, wherein one association word index element includes one association word and global attribute information of the multimedia resource capable of extracting the association word, and the server can accurately find the corresponding multimedia resource based on the global attribute information.
Step 209: updating corresponding association word index information in the global index library based on the one association word index element.
Specifically, the server updates the association word index information corresponding to the association word in the global index library based on the generated association word index element.
For example, an association word N exists in index information N of a real-time index library, the index information N corresponds to a multimedia resource N, corresponding association word N index information can be obtained by searching in a global index library based on the association word N, and the current association word index information includes: the association words n, the global attribute information of the multimedia resources A, the global attribute information of the multimedia files B and the global attribute information of the multimedia files C. When it is determined that the association word N is not recorded as a changed history association word, establishing an association word N index element based on the association word N and the global attribute information of the multimedia resource N, and updating corresponding association word N index information by using the association word N index element, that is, the updated association word N index information includes: the association word N, the global attribute information of the multimedia resource A, the global attribute information of the multimedia file B, the global attribute information of the multimedia file C and the global attribute information of the multimedia file N.
Step 210: and associating the association word with the global attribute information to generate an association word index element.
When the server determines that the read association word has no matched association word index information in the global index library, it is determined that the association word index information is not established based on the association word, so that an association word index element can be generated based on the association word and the global attribute information of the multimedia resource capable of reading the association word.
Step 211: and generating an association word index information in the global index library based on the association word index element.
The server generates an association word index element based on the read association word and the global attribute information of the multimedia resource which can read the association word in the real-time index library, and the global index library does not establish corresponding association word index information based on the association word, so that the server establishes the association word index information in the global index library directly based on the association word index element.
Step 212: is it determined whether all the associated words included in the index information have been read? If yes, go to step 201, otherwise go to step 204.
After the server completes updating and adding of the association word index information in the global index library based on the currently read association word, it further determines whether all the association words in the currently acquired index information are processed, if all the association words are completely read, then step 201 is further executed, the changed index information in the next index library is read, and the above operation is repeated. If not, the next associative word is continuously read, step 204 is executed, and the above operations are repeated.
Further, the method can be based on all multimedia resources in an offline state, wherein all multimedia resources are all multimedia resources which are obtained in real time before the current offline time point is reached, and only one multimedia resource which is obtained most recently is stored by one piece of ID information, so as to ensure the effectiveness of the stored multimedia resources. Performing offline batch processing calculation by using a Spark (Spark) cluster, wherein the calculation result can generate corresponding index information based on all multimedia resources, update the index information stored in a real-time index base, simultaneously, after extracting association words based on all multimedia resources, determine global attribute information of the multimedia resources capable of extracting the association words based on the extracted association words in the range of all multimedia resources, finally establish corresponding association word index information, and update the global index base based on the association word index information.
The following describes a process of recommending multimedia resources to an intelligent device by using the established global index library, with reference to fig. 3:
step 301: and responding to the association word query request sent by the intelligent equipment, and searching in the global index library.
The server responds to an association word query request sent by a user on the terminal device, and searches in a global index base, wherein the global index base is established based on index information in a real-time index base, the index information in the real-time index base is updated in real time based on the change of multimedia resource information in a network, one index information comprises at least one association word extracted from one multimedia resource and attribute information of the one multimedia resource, and the establishing processes of the real-time index base and the global index base are described in detail in the attached figures 1 and 2, and are not repeated herein.
Specifically, after acquiring the association words carried in the association word query request, the server searches a global index library to obtain association word index information matched with the association words, where the association word index information includes global attribute information of all multimedia resources from which the association words can be extracted, and the global attribute information at least includes device model information and authorization state information supported by corresponding multimedia resources.
Therefore, the global index database is updated in real time based on the updating of the real-time index database, so that the accuracy of the index information of the searched association words is ensured, the feasibility of operation is ensured, and the smooth operation of the whole recommendation process is ensured.
Step 302: and acquiring association word index information corresponding to the association word query request based on the search result, and acquiring corresponding multimedia resources based on the association word index information.
The server obtains association word index information corresponding to association words included in the association word query request based on search results, and then screens all the obtained multimedia resource information corresponding to the association words at least based on the equipment model of the intelligent equipment and the user identity identification information; screening out global attribute information of the multimedia resources matched with the equipment model of the intelligent equipment and the user identity identification information; and acquiring the multimedia resources corresponding to the global attribute information based on the screened global attribute information of the multimedia resources. Wherein the global attribute information at least includes: the language information of the multimedia resource, the media platform information to which the multimedia resource belongs, the terminal type information supported by the multimedia resource, the terminal model information supported by the multimedia resource, the service type information corresponding to the multimedia resource, and the ID information of the multimedia resource.
The server may also filter global attribute information of each multimedia resource associated with the associated word based on information of multiple dimensions, where the information of multiple dimensions may include: the service type information set by the intelligent equipment, the terminal type information to which the intelligent equipment belongs, the language information selected by the intelligent equipment and the like.
For example, suppose that a server receives an association word query request initiated by a user a on a terminal device a based on an association word a, the server may obtain that the query request includes a device signal Z of the terminal device a, a service type N set by the terminal device a, and a terminal type H to which the terminal device a belongs, and may determine that the user a is a member of a media platform M based on identity information of the user a. Further, the server searches in a global index library based on an association word a to obtain association word a index information matched with the association word a, the association word a index information includes global attribute information of a plurality of multimedia resources, the server screens the obtained global attribute information of the plurality of multimedia resources at least based on the identity information of the user a and a terminal type H to which a service type N terminal device a of a terminal device of which the device type Z terminal device is set belongs, and screens at least the device type Z-supporting media platform, which is a media platform M or another media platform that the user a can access and belongs to the service type N, and supports the global attribute information of the multimedia resources normally displayed by the terminal type H. And acquiring the multimedia resources corresponding to the global attribute information based on the global attribute information of the screened multimedia files.
Therefore, the multimedia resources which meet the requirements can be screened out based on the multi-dimensional requirements, on one hand, the effectiveness of normal display of the screened multimedia resources on the intelligent equipment can be guaranteed, on the other hand, the quality of the recommended multimedia resources can be guaranteed, and the situation of search errors can be avoided.
Step 303: and sending the multimedia resource to the intelligent equipment.
After the server screens out the multimedia resources, the obtained multimedia resources are sent to the intelligent equipment so that the intelligent equipment can further select and display the multimedia resources.
Based on the same inventive concept, fig. 4 is a schematic structural diagram illustrating an apparatus for recommending multimedia resources for updating suggested words in real time by using dual indexes, which is provided by the present application, and at least includes: search section 401, acquisition section 402, and transmission section 403, where:
the searching unit 401 searches in a global index base in response to an association word query request sent by an intelligent device, where the association word query request at least includes a device model of the intelligent device and user identification information, the global index base is established based on index information in a real-time index base, the index information in the real-time index base is updated in real time based on changes of multimedia resource information in a network, and one index information includes at least one association word extracted from one multimedia resource and attribute information of the one multimedia resource.
An obtaining unit 402, configured to obtain, based on a search result, association word index information corresponding to the association word query request, and obtain a corresponding multimedia resource based on the association word index information.
A sending unit 403, sending the multimedia resource to the intelligent device.
Based on the same inventive concept, fig. 5 exemplarily illustrates a schematic structural diagram of a computing device provided in an embodiment of the present application, and includes at least a memory 501 and a processor 502;
a memory 501 for storing program instructions;
the processor 502 is used for calling the program instructions stored in the memory and executing the method according to the obtained program.
Based on the same inventive concept, the embodiment of the present invention also provides a computer-readable non-volatile storage medium, which includes computer-readable instructions, and when the computer-readable instructions are read and executed by a computer, the computer-readable instructions cause the computer to execute the above information query method.
In summary, in the present application, in response to an association word query request sent by an intelligent device, a search is performed in a global index library, the global index library is established based on index information in a real-time index library, the index information in the real-time index library is updated in real time based on changes of multimedia resource information in a network, then, based on a search result, association word index information corresponding to the association word query request is obtained, a corresponding multimedia resource is obtained based on the association word index information, and then, the multimedia resource is sent to the intelligent device. Therefore, the consistency between the actual attribute information of the multimedia resources and the attribute information recorded in the index library is ensured based on the global index library updated in real time, the accuracy of the multimedia resources obtained by the intelligent equipment is also ensured, the processing efficiency is improved, and the user experience is improved.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
While preferred embodiments of the present invention have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. Therefore, it is intended that the appended claims be interpreted as including preferred embodiments and all such alterations and modifications as fall within the scope of the invention.
It will be apparent to those skilled in the art that various modifications and variations can be made in the embodiments of the present invention without departing from the spirit or scope of the embodiments of the invention. Thus, if such modifications and variations of the embodiments of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to encompass such modifications and variations.
Claims (10)
1. A multimedia resource recommendation method for updating associative words in real time through double indexes is characterized by comprising the following steps:
responding to an association word query request sent by intelligent equipment, searching in a global index library, wherein the association word query request at least comprises an equipment model of the intelligent equipment and user identity identification information, the global index library is established based on index information in a real-time index library, the index information in the real-time index library is updated in real time based on the change of multimedia resource information in a network, and one piece of index information comprises at least one association word extracted from one multimedia resource and attribute information of the one multimedia resource;
acquiring association word index information corresponding to the association word query request based on a search result, and acquiring corresponding multimedia resources based on the association word index information;
and sending the multimedia resource to the intelligent equipment.
2. The method of claim 1, wherein the responding to the request for the query for the associative word sent by the smart device is preceded by: establishing a real-time index library; the method specifically comprises the following steps:
establishing a real-time index library based on attribute information of multimedia resources and association words extracted from titles or labels of the multimedia resources;
reading the changed multimedia resources in the network in real time, and executing the following operations for each read multimedia resource:
extracting at least one association word contained in a multimedia resource, acquiring ID information contained in attribute information of the multimedia resource, and obtaining an information digest key based on the attribute information of the multimedia resource;
generating index information based on the at least one association word and attribute information of the multimedia asset when it is determined that the multimedia asset is not recorded in a real-time index repository based on the ID information;
and storing the index information to the real-time index library.
3. The method according to claim 2, wherein after extracting at least one associative word included in the multimedia asset, obtaining ID information included in attribute information of the multimedia asset, and obtaining information digest key information based on the attribute information of the multimedia asset, the method further comprises:
when it is determined that the multimedia asset is recorded in a real-time index repository based on the ID information, determining index information corresponding to the multimedia asset in the real-time index repository based on the ID information;
acquiring a historical information abstract key of the multimedia resource and at least one historical association word included in the index information;
and when the information digest key is determined to be different from the historical information digest key and the at least one association word is determined to be the same as the at least one historical association word, updating the index information based on the attribute information of the multimedia resource.
4. The method of claim 3, wherein after obtaining the historical information digest key value of the multimedia resource and the at least one historical association word included in the index information, further comprising:
and when the information digest key is determined to be different from the historical information digest key and the at least one association word is determined to be different from the at least one historical association word, recording the at least one historical association word, and updating the index information based on the at least one association word and the attribute information.
5. The method of any one of claims 2-4, wherein prior to responding to the request for the query for the associative word sent by the smart device, further comprising: establishing a global index library; the method specifically comprises the following steps:
establishing a global index library based on the association words and the global attribute information of the associated multimedia resources;
acquiring the index information of each change in the real-time index library in real time, and respectively executing the following operations:
extracting at least one association word contained in the index information and attribute information of the associated multimedia resource, and determining global attribute information of the multimedia resource based on the attribute information of the multimedia resource;
when it is determined that the at least one association word does not have matched association word index information in the global index library, associating the at least one association word with the global attribute information respectively to generate at least one association word index element;
generating at least one piece of associative word index information in the global index repository based on the at least one associative word index element.
6. The method of claim 5, wherein after extracting at least one of the associative word and the attribute information of the associated multimedia resource from the index information and determining the global attribute information of the multimedia resource based on the attribute information of the multimedia resource, further comprising:
when determining that the at least one associated word has matching associated word index information in the global index library, respectively performing the following operations on each associated word in the at least one associated word:
and when the association word is the historical association word recorded in the index information, updating the association word index information in a global index library based on the association word.
7. The method of claim 5, wherein after extracting at least one of the associative word and the attribute information of the associated multimedia resource from the index information and determining the global attribute information of the multimedia resource based on the attribute information of the multimedia resource, further comprising:
when determining that the at least one associated word has matching associated word index information in the global index library, respectively performing the following operations on each associated word in the at least one associated word:
when the association word is not the historical association word recorded in the index information, associating the association word with the global attribute information to generate an association word index element;
updating the associative word index information based on the associative word index element.
8. The method of claim 1, wherein the searching in the global index repository comprises:
acquiring the association words carried in the association word query request;
searching a global index library to obtain association word index information matched with the association words, wherein the association word index information comprises global attribute information of all multimedia resources capable of extracting the association words, and the global attribute information at least comprises equipment model information and authorization state information supported by the corresponding multimedia resources.
9. The method of claim 4, wherein obtaining the associative word index information corresponding to the associative word query request comprises:
screening all the obtained multimedia resource information corresponding to the association words at least based on the equipment model of the intelligent equipment and the user identity identification information;
screening out global attribute information of the multimedia resources matched with the equipment model of the intelligent equipment and the user identity identification information;
and acquiring the multimedia resources corresponding to the global attribute information based on the screened global attribute information of the multimedia resources.
10. A multimedia resource recommendation device for updating associative words in real time through double indexes is characterized by comprising the following steps:
the searching unit is used for responding to an association word query request sent by intelligent equipment and searching in a global index library, wherein the association word query request at least comprises the equipment model of the intelligent equipment and user identity identification information, the global index library is established based on index information in a real-time index library, the index information in the real-time index library is updated in real time based on the change of multimedia resource information in a network, and one piece of index information comprises at least one association word extracted from one multimedia resource and attribute information of the one multimedia resource;
the acquisition unit is used for acquiring association word index information corresponding to the association word query request based on a search result and acquiring corresponding multimedia resources based on the association word index information;
and the sending unit is used for sending the multimedia resource to the intelligent equipment.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911228998.XA CN111143582B (en) | 2019-12-04 | 2019-12-04 | Multimedia resource recommendation method and device for updating association words in double indexes in real time |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911228998.XA CN111143582B (en) | 2019-12-04 | 2019-12-04 | Multimedia resource recommendation method and device for updating association words in double indexes in real time |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111143582A true CN111143582A (en) | 2020-05-12 |
| CN111143582B CN111143582B (en) | 2023-09-22 |
Family
ID=70517557
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911228998.XA Active CN111143582B (en) | 2019-12-04 | 2019-12-04 | Multimedia resource recommendation method and device for updating association words in double indexes in real time |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111143582B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115314737A (en) * | 2021-05-06 | 2022-11-08 | 青岛聚看云科技有限公司 | Content display method, display equipment and server |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101072205A (en) * | 2007-06-21 | 2007-11-14 | 腾讯科技（深圳）有限公司 | Chat information searching method and system |
| CN103218364A (en) * | 2012-01-19 | 2013-07-24 | 阿里巴巴集团控股有限公司 | Searching method and system |
| CN104778267A (en) * | 2015-04-22 | 2015-07-15 | 无锡天脉聚源传媒科技有限公司 | Searching and index updating method and device |
| CN106294768A (en) * | 2016-08-11 | 2017-01-04 | 深圳市宜搜科技发展有限公司 | Information search method and information search engine |
| CN107943893A (en) * | 2017-11-16 | 2018-04-20 | 北京奇安信科技有限公司 | A kind of search processing method and device based on internet |
-
2019
- 2019-12-04 CN CN201911228998.XA patent/CN111143582B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101072205A (en) * | 2007-06-21 | 2007-11-14 | 腾讯科技（深圳）有限公司 | Chat information searching method and system |
| CN103218364A (en) * | 2012-01-19 | 2013-07-24 | 阿里巴巴集团控股有限公司 | Searching method and system |
| CN104778267A (en) * | 2015-04-22 | 2015-07-15 | 无锡天脉聚源传媒科技有限公司 | Searching and index updating method and device |
| CN106294768A (en) * | 2016-08-11 | 2017-01-04 | 深圳市宜搜科技发展有限公司 | Information search method and information search engine |
| CN107943893A (en) * | 2017-11-16 | 2018-04-20 | 北京奇安信科技有限公司 | A kind of search processing method and device based on internet |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115314737A (en) * | 2021-05-06 | 2022-11-08 | 青岛聚看云科技有限公司 | Content display method, display equipment and server |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111143582B (en) | 2023-09-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20200257543A1 (en) | Aggregate Features For Machine Learning | |
| RU2501078C2 (en) | Ranking search results using edit distance and document information | |
| CN102156751B (en) | Method and device for extracting video fingerprint | |
| US20120117051A1 (en) | Multi-modal approach to search query input | |
| CN109408821B (en) | Corpus generation method and device, computing equipment and storage medium | |
| CN110909182A (en) | Multimedia resource searching method and device, computer equipment and storage medium | |
| CN111400586A (en) | Group display method, terminal, server, system and storage medium | |
| US20110179013A1 (en) | Search Log Online Analytic Processing | |
| CN108509545B (en) | Method and system for processing comments of article | |
| CN111652658A (en) | Portrait fusion method, apparatus, electronic device and computer readable storage medium | |
| KR102575507B1 (en) | Article writing soulution using artificial intelligence and device using the same | |
| CN111368100A (en) | Media asset merging method and device thereof | |
| CN111143582B (en) | Multimedia resource recommendation method and device for updating association words in double indexes in real time | |
| CN111428120B (en) | Information determination method and device, electronic equipment and storage medium | |
| CN110765348B (en) | Hot word recommendation method and device, electronic equipment and storage medium | |
| CN104376000A (en) | Webpage attribute determination method and webpage attribute determination device | |
| CN110569447A (en) | network resource recommendation method and device and storage medium | |
| CN112749296A (en) | Video recommendation method and device, server and storage medium | |
| CN115729965A (en) | Information stream processing method, device, stream server and storage medium | |
| CN112766779B (en) | Information processing method, computer device, and storage medium | |
| JP2019200582A (en) | Search device, search method, and search program | |
| CN110929207B (en) | Data processing method, device and computer readable storage medium | |
| CN114564501A (en) | Database data storage and query methods, devices, equipment and medium | |
| CN114357242A (en) | Training evaluation method and device based on recall model, equipment and storage medium | |
| CN107025615B (en) | Learning condition statistical method based on learning tracking model |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |