CN103491205A - Related resource address push method and device based on video retrieval - Google Patents

Abstract

The invention discloses a method for pushing associated resource addresses based on video search, which includes: when receiving a loading or playing request of the first video resource data, acquiring feature text information of the first video resource data; The text information of the above features is mapped to one or more first participle; find the associated second participle whose co-occurrence rate with the one or more first participle is higher than the preset threshold; the co-occurrence rate of the current one or more The probability that a plurality of first participle and the second participle co-occur in the same video resource data; obtain the network link address of the second video resource data matching the one or more first participle and the associated second participle; Pushing the network link address of the second video resource data. The invention realizes deep excavation of high-quality resources in the video library, and improves the efficiency of resource excavation; in addition, the index table will continuously expand with the continuous accumulation of Internet video content, which is conducive to expanding the recall rate.

Description

A method and device for pushing associated resource addresses based on video search

technical field

The present invention relates to the technical field of the Internet, in particular to a method for pushing associated resource addresses based on video search and a device for pushing associated resource addresses based on video search.

Background technique

Video search engine is a vertical search technology different from comprehensive search. A video search engine crawls and indexes video results on the Internet, and since it can provide pure video results to searchers, it can greatly save Internet users' time in searching for videos.

According to the relevant statistical data of video search, entertainment, games, film and television, news, animation and other types of videos are the main search objects of users. This shows that users have a general demand for video search itself. Users often do not have a strong purpose, and the search results are not "necessary", but have a certain degree of scalability, as long as the target is within the user's favorite category. Therefore, relevant recommendations are often made to users outside of search results.

However, the existing video search engines are not enough in terms of relevant recommendations: some video search engines do not have relevant recommendations, and video search engines with relevant recommendations are only based on the user's search history data, through manual sorting to obtain the correlation system and other simple methods Implement referrals. This kind of recommendation system is based on the user's existing search habits, and the recall rate is low. In addition, because the user's search range is generally much smaller than the resource range in the existing Internet, it cannot fully mine high-quality videos in the Internet.

Another search recommendation method is to rely on manually sorting out a resource association system or obtain such a system from other knowledge systems, and apply it to the recommendation system. For example, when you search for "square dance" on a certain search engine, you will get recommended words such as "ballroom dance", "belly dance", and "aerobics", and when you search for "dota", you will get "Cross Fire", "World of Warcraft", etc. recommended words, but the recall rate of this system is low, and generally cannot give recommendations in long-tail searches.

Contents of the invention

In view of the above problems, the present invention is proposed to provide a method for pushing associated resource addresses based on video search and a corresponding device for pushing associated resource addresses based on video search, which overcome the above problems or at least partially solve the above problems .

According to one aspect of the present invention, a method for pushing associated resource addresses based on video search is provided, including:

When receiving the loading or playing request of the first video resource data, acquire the feature text information of the first video resource data;

Mapping the feature text information into one or more first participles;

Finding the associated second participle whose co-occurrence rate with the one or more first participle is higher than the preset threshold; the co-occurrence rate of the current one or more first participle and the second participle are in the same video resource data probability of co-occurrence;

Acquire the network link address of the second video resource data matching the one or more first participle and the associated second participle;

Pushing the network link address of the second video resource data.

Optionally, when the loading or playing request of the first video resource data is received, the step of acquiring the feature text information of the first video resource data includes:

When receiving the play request of the first video data, receiving the characteristic text information of the first video resource data sent by the current terminal;

or,

When the first video data loading request is received, the feature text information of the video resource data preset locally is extracted.

Optionally, the step of mapping the feature text information into one or more first word segmentations includes:

Extracting a participle mapped by the text information of the feature book;

or,

When the received feature text information is a compound word, split the feature text information into multiple search subwords; and extract multiple word segmentations mapped to the multiple search subwords.

Optionally, the step of searching for the associated second participle whose co-occurrence rate with the one or more first participle is higher than a preset threshold includes:

When the feature text information is mapped to a first participle, extract a preset index table corresponding to the first participle; wherein, the index table includes information about the video resource data to which the first participle belongs, and , all the word segmentations in the video resource data; all the word segmentations in the video resource data are generated by capturing the video resource data, extracting the feature text information of the video resource data, and performing word segmentation on the feature text information;

Calculate the co-occurrence rate of the first participle and each second participle in the index table, the co-occurrence rate is the number of occurrences of each second participle in the index table and the information of the video resource data in the index table The ratio of the total number; wherein, the second participle is a participle except the first participle among all participles in the video resource data;

Extracting the second participle whose co-occurrence rate is higher than the preset threshold as the associated second participle.

Optionally, the step of searching for the associated second participle whose co-occurrence rate with the one or more first participle is higher than a preset threshold includes:

When the feature text information is mapped to a plurality of first participles, a plurality of preset index tables corresponding to the plurality of first participles are respectively extracted; each index table includes the video resource data to which the first participle belongs information, and all the word segmentation in the video resource data; all the word segmentation in the video resource data is to extract the feature text information of the video resource data by grabbing the video resource data, and carry out the feature text information on the feature text information participle generation;

Extracting a second word segment that co-occurs with the plurality of first word segments as a candidate word segment; wherein, the second word segment is a word segment in all word segments in the video resource data except the first word segment;

Calculate the co-occurrence rate of the first participle and the candidate participle in each index table respectively, the co-occurrence rate is the number of occurrences of the candidate participle in the index table and the total information of the video resource data in the index table ratio of

Configure a plurality of weights corresponding to the co-occurrence ratios of the plurality of first participle and the candidate participle respectively;

Calculating the average of multiple weighted co-occurrence rates, respectively, as the co-occurrence rates of the multiple first participle and the candidate participle;

Extracting the candidate participle whose co-occurrence rate is higher than the preset threshold as the associated second participle.

Optionally, the step of searching for the associated second participle whose co-occurrence rate with the one or more first participle is higher than a preset threshold includes:

When the feature text information is mapped to a plurality of first participles, a plurality of preset index tables corresponding to the plurality of first participles are respectively extracted; wherein, each index table includes the video to which the first participle belongs The information of the resource data, and all word segmentations in the video resource data; all the word segmentations in the video resource data are to extract the feature text information of the video resource data by grabbing the video resource data, and to the feature text information for word segmentation;

Using the plurality of index tables to determine the main participle, the main participle is the first participle corresponding to the index table with the largest total information of the video resource data;

Calculate the co-occurrence rate of each second participle in the main participle and its corresponding index table, the co-occurrence rate is the number of occurrences of each second participle in the index table and the total number of video resource data information in the index table ratio; wherein, the second participle is a participle except the first participle among all participles in the video resource data;

Extracting the second participle whose co-occurrence rate is higher than the preset threshold as the associated second participle.

Optionally, the feature text information includes video title, video keywords and/or video description.

Optionally, the step of obtaining the network link address of the second video resource data matching the one or more first participle and the associated second participle includes:

Obtaining the network link addresses of the main participle and the second video resource data associated with the second participle.

According to another aspect of the present invention, a device for pushing associated resource addresses based on video search is provided, including:

The characteristic text information acquisition module is adapted to obtain the characteristic text information of the first video resource data when receiving the loading or playing request of the first video resource data;

The first participle mapping module is adapted to map the feature text information into one or more first participles;

The second participle search module is adapted to search for the associated second participle whose co-occurrence rate with the one or more first participle is higher than the preset threshold; the co-occurrence rate of the current one or more first participle and the second The probability that the dichotomous words co-occur in the same video resource data;

A network connection address obtaining module, adapted to obtain the network link address of the second video resource data matching the one or more first participle and the associated second participle;

The network connection address pushing module is adapted to push the network link address of the second video resource data.

Optionally, the feature text information acquisition module is also suitable for:

When receiving the play request of the first video data, receiving the characteristic text information of the first video resource data sent by the current terminal;

or,

When the first video data loading request is received, the feature text information of the video resource data preset locally is extracted.

Optionally, the first word segmentation mapping module is also suitable for:

Extracting a participle mapped by the text information of the feature book;

or,

When the received feature text information is a compound word, split the feature text information into multiple search subwords; and extract multiple word segmentations mapped to the multiple search subwords.

Optionally, the second word segmentation search module is also suitable for:

When the feature text information is mapped to a first participle, extract a preset index table corresponding to the first participle; wherein, the index table includes information about the video resource data to which the first participle belongs, and , all the word segmentations in the video resource data; all the word segmentations in the video resource data are generated by capturing the video resource data, extracting the feature text information of the video resource data, and performing word segmentation on the feature text information;

Calculate the co-occurrence rate of the first participle and each second participle in the index table, the co-occurrence rate is the number of occurrences of each second participle in the index table and the information of the video resource data in the index table The ratio of the total number; wherein, the second participle is a participle except the first participle among all participles in the video resource data;

Extracting the second participle whose co-occurrence rate is higher than the preset threshold as the associated second participle.

Optionally, the second word segmentation search module is also suitable for:

When the feature text information is mapped to a plurality of first participles, a plurality of preset index tables corresponding to the plurality of first participles are respectively extracted; each index table includes the video resource data to which the first participle belongs information, and all the word segmentation in the video resource data; all the word segmentation in the video resource data is to extract the feature text information of the video resource data by grabbing the video resource data, and carry out the feature text information on the feature text information participle generation;

Extracting a second word segment that co-occurs with the plurality of first word segments as a candidate word segment; wherein, the second word segment is a word segment in all word segments in the video resource data except the first word segment;

Calculate the co-occurrence rate of the first participle and the candidate participle in each index table respectively, the co-occurrence rate is the number of occurrences of the candidate participle in the index table and the total information of the video resource data in the index table ratio of

Configure a plurality of weights corresponding to the co-occurrence ratios of the plurality of first participle and the candidate participle respectively;

Calculating the average of multiple weighted co-occurrence rates, respectively, as the co-occurrence rates of the multiple first participle and the candidate participle;

Extracting the candidate participle whose co-occurrence rate is higher than the preset threshold as the associated second participle.

Optionally, the second word segmentation search module is also suitable for:

When the feature text information is mapped to a plurality of first participles, a plurality of preset index tables corresponding to the plurality of first participles are respectively extracted; wherein, each index table includes the video to which the first participle belongs The information of the resource data, and all word segmentations in the video resource data; all the word segmentations in the video resource data are to extract the feature text information of the video resource data by grabbing the video resource data, and to the feature text information for word segmentation;

Using the plurality of index tables to determine the main participle, the main participle is the first participle corresponding to the index table with the largest total information of the video resource data;

Calculate the co-occurrence rate of each second participle in the main participle and its corresponding index table, the co-occurrence rate is the number of occurrences of each second participle in the index table and the total number of video resource data information in the index table ratio; wherein, the second participle is a participle except the first participle among all participles in the video resource data;

Extracting the second participle whose co-occurrence rate is higher than the preset threshold as the associated second participle.

Optionally, the feature text information includes video title, video keywords and/or video description.

Optionally, the network connection address acquisition module is also suitable for:

Obtain the network link addresses of the main participle and the second video resource data associated with the second participle.

The present invention can push according to the existing released content, so that the search engine can get rid of the dependence on the user's search habit, and push out the video resource data that although few users search, but the video library has more related resources, so as to realize In-depth excavation of high-quality resources in the video library improves the efficiency of resource mining; in addition, the index table will continue to expand with the continuous accumulation of Internet video content, and the quantity and breadth of content produced by major video stations will far exceed that of users. The number of searched words is conducive to expanding the recall rate.

In the present invention, by obtaining the network connection address of the second video resource data matched by the first participle and the second participle, the user can directly obtain the video data resource based on this address, so that the user can obtain more results by simply searching without The search is submitted multiple times, thereby reducing the burden of accessing the server, reducing the occupation of network resources, and improving user experience.

The above description is only an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention, it can be implemented according to the contents of the description, and in order to make the above and other purposes, features and advantages of the present invention more obvious and understandable , the specific embodiments of the present invention are enumerated below.

Description of drawings

Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiment. The drawings are only for the purpose of illustrating a preferred embodiment and are not to be considered as limiting the invention. Also throughout the drawings, the same reference numerals are used to designate the same parts. In the attached picture:

FIG. 1 shows a flow chart of the steps of an embodiment of a method for pushing an associated resource address based on a video search according to an embodiment of the present invention; and

Fig. 2 shows a structural block diagram of an embodiment of an apparatus for pushing associated resource addresses based on video search according to an embodiment of the present invention.

Detailed ways

Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided for more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

Referring to FIG. 1 , it shows a flow chart of the steps of an embodiment of pushing an associated resource address based on a video search according to an embodiment of the present invention, which may specifically include the following steps:

Step 101, when a loading or playing request of the first video resource data is received, acquire feature text information of the first video resource data;

It should be noted that the first video resource data may be located on the terminal device or on the network, and the feature text information may be the information carried by the video resource data.

In a preferred embodiment of the present invention, the step 101 may specifically include the following sub-steps:

Sub-step S11, when receiving the play request of the first video data, receiving the characteristic text information of the first video resource data sent by the current terminal;

When the first video resource data is located on the terminal device, the characteristic text information of the first video resource data may be extracted by the terminal device, and then uploaded to the corresponding server side.

or,

Sub-step S12, when receiving the first video data loading request, extracting the locally preset feature text information of the video resource data.

When the first video resource data is located on the network, the feature text information of the first video resource data may be extracted by the server side.

In a preferred embodiment of the present invention, the feature text information may include video title, video keywords and/or video description.

For example, in a piece of video resource data titled "[Photographer] Dongguan became Venice after a heavy rain, more than a thousand cars flooded and broke down—online broadcast—XX.com, video HD online viewing", the characteristic text information can be as follows:

Title of the video (Title): [Photographer] After the rainstorm in Dongguan, it turned into Venice, and more than a thousand cars were flooded and broke down—Online Play—XX Network, watch the video online in high-definition;

Video keywords (Keywords): YY reporter's life information Dongguan flooding;

Video description (Description): A heavy rain yesterday morning made the neighbors in some parts of Dongguan feel as if they had come to Venice. The driving car was flooded and broke down in the heavy rain, and some neighbors' homes were also covered with ocean.

In practical applications, the feature text information can be a word, that is, include a semantically independent word, such as Mid-Autumn Festival, Dragon Boat Festival, National Day, etc.; the feature text information can also be a compound word, that is, include two or more semantically independent words. Words, such as Mid-Autumn Festival moon cakes, Dragon Boat Festival rice dumplings, National Day Tibet travel and so on. Generally speaking, the video resource data in the terminal device often only has the video title (Title), such as "Iron Man", "Spiderman" and other movie titles; the video resource data in the network often includes the video title (Title), video One or more of keywords (Keywords) and video description (Description).

Step 102, mapping the feature text information into one or more first participles;

It should be noted that the mapped word segments can be preset and can be used to calculate the co-occurrence rate between different word segments.

The mapping rules can also be one or more preset, which can include words that have no practical meaning such as removing dirty words, modifiers, modal particles, broad words, etc. of video search characters; can include setting stop words, that is, some common Words are the criteria for stopping when splitting phrases, such as, I, you, etc.; it can also include the correspondence of association relations, corresponding to multiple expressions of the same thing as one expression, for example, August 15th, Mid-Autumn Festival , Mooncake Festival, etc. are associated with Mid-Autumn Festival; other mapping rules may also be included, which is not limited in this embodiment of the present invention.

English is based on words, and words are separated by spaces, while Chinese is based on words, and all the words in a sentence can be connected to describe a meaning. For example, the English sentence I ama student, in Chinese it is: "I am a student". The computer can easily know that student is a word through the blank space, but it cannot easily understand that the combination of the words "learning" and "sheng" means a word. Segmenting Chinese character sequences into meaningful words is Chinese word segmentation. For example, I am a student, the word segmentation result is: I, is, one, student.

Here are some common word segmentation methods:

1. Word segmentation method based on string matching: it refers to matching the Chinese character string to be analyzed with the entry in a preset machine dictionary according to a certain strategy. If a certain string is found in the dictionary, the match is successful ( recognized a word). The actual word segmentation system uses mechanical word segmentation as a means of initial segmentation, and it is necessary to use various other language information to further improve the accuracy of segmentation.

2. Word segmentation method based on feature scanning or token segmentation: it refers to identifying and segmenting some words with obvious characteristics in the character string to be analyzed first, and using these words as breakpoints, the original character string can be divided into relatively Small strings are then mechanically segmented to reduce the matching error rate; or combine word segmentation and part-of-speech tagging, use rich part-of-speech information to help word segmentation decisions, and in turn check and adjust the word segmentation results during the tagging process , so as to improve the accuracy of segmentation.

3. Comprehension-based word segmentation method: It refers to the effect of recognizing words by letting the computer simulate the human understanding of the sentence. Its basic idea is to perform syntactic and semantic analysis at the same time of word segmentation, and use syntactic information and semantic information to deal with ambiguity. It usually includes three parts: the word segmentation subsystem, the syntax and semantics subsystem, and the general control part. Under the coordination of the general control part, the word segmentation subsystem can obtain syntactic and semantic information about words and sentences to judge the ambiguity of word segmentation, that is, it simulates the process of human understanding of sentences. This word segmentation method requires the use of a large amount of language knowledge and information.

4. The word segmentation method based on statistics: it means that the frequency or probability of adjacent co-occurrence of words in Chinese information can better reflect the credibility of the word, so it is possible to analyze the adjacent co-occurrence of each word in the corpus. Combination frequency statistics, calculate their mutual occurrence information, and calculate the adjacent co-occurrence probability of two Chinese characters X, Y. Mutual appearance information can reflect the closeness of the combination relationship between Chinese characters. When the degree of closeness is higher than a certain threshold, it can be considered that this word group may form a word. This method only needs to count the frequency of words in the corpus, and does not need to divide the dictionary.

In a preferred embodiment of the present invention, the step 102 may specifically include the following sub-steps:

Sub-step S21, extracting a word segment mapped to the feature book text information;

For the case where the feature text information is a word, its corresponding word segmentation can be directly extracted according to the preset mapping rules. For example, if the feature text information is "Mid-Autumn Festival", "My Mid-Autumn Festival" or "Mid-Autumn Festival", etc., the first participle of the mapping may be "Mid-Autumn Festival". Of course, the text information of the feature text and the first participle mapped to it can also be the same word. For example, the text information of the feature text is "Mid-Autumn Festival", and the mapped first participle can also be "Mid-Autumn Festival".

or,

Sub-step S22, when the received feature text information is a compound word, split the feature text information into multiple search subwords;

Sub-step S23, extracting multiple word segmentations mapped to the multiple search subwords.

For the case where the feature text information is a compound word, the word segmentation can be performed according to the preset mapping rules to obtain the search subwords, and then the word segmentation corresponding to the search subwords can be extracted respectively. For example, if the received feature text information is "Mid-Autumn Festival Mooncake", it can be split into two search subwords "Mid-Autumn Festival" and "Mooncake", and then "Mid-Autumn Festival" is mapped to "Mid-Autumn Festival", and "Mid-Autumn Festival" is mapped to "Mid-Autumn Festival" "Moon cake" is mapped to "moon cake", and two first participles of "Mid-Autumn Festival" and "moon cake" are obtained.

Step 103, searching for associated second participle whose co-occurrence rate with the one or more first participle is higher than a preset threshold;

The co-occurrence rate is the probability that the current one or more first participle and the second participle co-occur in the same video resource data;

Specifically, the co-occurrence rate may be the probability that one or more current participle and the second participle co-occur in the characteristic text information of the same video resource data, specifically, it may include a co-occurrence rate of the first participle and the second participle, The co-occurrence rate of multiple participle and second participle.

It should be noted that, the second participle may be a participle other than the first participle among all the preset participles. The associated second participle may be a second participle whose co-occurrence rate with the first participle is higher than a preset threshold.

In practical applications, the video resource data may include feature text information, and the feature text information may be used to record related information of the video resource data, and may also be used to extract word segmentation.

In a preferred embodiment of the present invention, the step 103 may specifically include the following sub-steps:

Sub-step S31, when the feature text information is mapped to a first participle, extract the preset index table corresponding to the first participle; wherein, the index table includes the video resource data to which the first participle belongs information, and all the word segmentation in the video resource data; all the word segmentation in the video resource data is to extract the feature text information of the video resource data by grabbing the video resource data, and carry out the feature text information on the feature text information participle generation;

In the specific implementation, the search engine can be used in advance to grab the video resource data on each website platform through crawlers, and then build an index library: extract the characteristic text information of the video resource data for word segmentation processing, and create an index table corresponding to each word segment, The index table can store the information of video resource data (it can be ID, internal network address, external network address, etc. video identifiers, or a record composed of current word segmentation and other word segmentation), all information in video resource data participle (including the first participle and the second participle except the first participle).

In a preferred embodiment of the present invention, the feature text information may include video title, video keywords and/or video description.

For example, the index table for "Mid-Autumn Festival" can be as follows:

Wherein, the first participle is "Mid-Autumn Festival", and the information of the video resource data includes the video identification. Of course, the information of the video resource data may not include the video identifier, but only the record formed by the first participle and the second participle (that is, the second participle in each line is regarded as a record).

Certainly, the above index table is only an example, and other index tables may be set according to actual conditions when implementing the embodiment of the present invention, which is not limited in the embodiment of the present invention. In addition, in addition to the above index tables, those skilled in the art may also use other index tables according to actual needs, which is not limited in this embodiment of the present invention.

It should be noted that the video resource data on each platform can be captured periodically or irregularly, and then the index can be updated to build a database, that is, each index table can be updated.

Sub-step S32, calculating the co-occurrence rate of the first participle and each second participle in the index table, the co-occurrence rate is the same as the number of occurrences of each second participle in the index table and the video frequency in the index table The ratio of the total information of the resource data; wherein, the second participle is a participle except the first participle among all participles in the video resource data;

Since the number of occurrences of each second participle in the index table is the same as the number of video data data it belongs to, the co-occurrence rate can also be expressed as the same as the number of video data data that each second participle belongs to in the index table and in the index table The ratio of the total information of video resource data.

For example, there are a total of 100 pieces of video resource data information in the index table of the word segmentation "square dance", and a total of 200 video resource data information in the index table of the word segmentation "Bing Ge". There are 10 pieces of video resource data information appearing in these two index tables, then for "square dance", the co-occurrence rate of "square dance" and "Binggege" is 10/100=10%, and for For "Bing Ge", the co-occurrence rate of "Bing Ge" and "Square Dance" is 10/200=5%.

Sub-step S33, extracting the second participle whose co-occurrence rate is higher than the preset threshold as the associated second participle.

In a specific implementation, the preset threshold may be set by those skilled in the art according to actual conditions, which is not limited in this embodiment of the present invention. The associated second participle extracted in the embodiment of the present invention may be empty, or may be one or more.

In a preferred embodiment of the present invention, the step 103 may specifically include the following sub-steps:

Sub-step S41, when the feature text information is mapped to a plurality of first participles, respectively extract a plurality of preset index tables corresponding to the plurality of first participles; each index table includes the information of the video resource data, and all the word segmentation in the video resource data; all the word segmentation in the video resource data is to extract the feature text information of the video resource data by grabbing the video resource data, and to the described feature text information for word segmentation;

In the specific implementation, the search engine can be used in advance to grab the video resource data on each platform through the crawler, and then build an index and build a database: extract the characteristic text information of the video resource data for word segmentation processing, and establish an index table corresponding to each word segment, The index table can store the information of video resource data (it can be ID, internal network address, external network address, etc. video identifiers, or a record composed of current word segmentation and other word segmentation), all information in video resource data participle (including the first participle and the second participle except the first participle).

In a preferred embodiment of the present invention, the feature text information may include video title, video keywords and/or video description.

Sub-step S42, extracting a second participle that co-occurs with the plurality of first participle as a candidate participle; wherein, the second participle is a participle other than the first participle among all participles in the video resource data ;

Specifically, there are currently multiple first participles, that is, there are multiple index tables corresponding to the number, and the candidate participle needs to appear in each index table, that is, the candidate participle and the current first participle all appear in the same index table .

Sub-step S43, respectively calculate the co-occurrence rate of the first participle and the candidate participle in each index table, the co-occurrence rate is the number of occurrences of the candidate participle in the index table and the video resource in the index table The ratio of the total number of information in the data;

For example, the feature text information "Mid-Autumn Festival Mooncake" can be mapped to the first participle "Mid-Autumn Festival" and "Mooncake", and one of the candidate participle is extracted as "Moon", then the simultaneous calculation of "Mid-Autumn Festival" and "Moon" can be calculated respectively. The occurrence rate (assumed to be 70%), the co-occurrence rate of "mooncake" and "moon" (assumed to be 60%).

Sub-step S44, respectively configuring a plurality of weights corresponding to the co-occurrence ratios of the plurality of first participle and the candidate participle;

The weight can be determined according to the proportion of the total information of video resource data in the index table between each first participle, wherein the more the total information of video resource data in the index table, the greater the weight. For example, if the total number of video resource data information in the "Mid-Autumn" index table is 900, and the total number of video resource data information in the "Mooncake" index table is 100, then the co-occurrence rate of "Mid-Autumn Festival" and "Moon" is The weight can be 0.9, and the weight of the co-occurrence rate of "moon cake" and "moon" can be 0.1.

Of course, the above weights are only examples. When implementing the embodiment of the present invention, other weights can be set according to actual conditions, for example, setting corresponding weights according to current social hotspots (news rankings, Weibo rankings, etc.), according to the user's local and/or The corresponding weights and the like are set for online operation behaviors (video playing, news reading, etc.), which are not limited in this embodiment of the present invention. In addition, in addition to the above weights, those skilled in the art may also adopt other weights according to actual needs, which is not limited in this embodiment of the present invention.

Sub-step S45, respectively calculating the average value of multiple weighted co-occurrence rates as the co-occurrence rates of the multiple first participle and the candidate participle;

In the embodiment of the present invention, the weighted average of multiple co-occurrence rates may be used as the final co-occurrence rate.

For example, the co-occurrence rate of "Mid-Autumn Festival", "Mooncake" and "Moon" can be (70%*0.9+60%*0.1)/2=34.5%.

Sub-step S46, extracting the candidate participle whose co-occurrence rate is higher than the preset threshold as the associated second participle.

In a specific implementation, the preset threshold may be set by those skilled in the art according to actual conditions, which is not limited in this embodiment of the present invention. The associated second participle extracted in the embodiment of the present invention may be empty, or may be one or more.

In a preferred embodiment of the present invention, the step 103 may specifically include the following sub-steps:

Sub-step S51, when the feature text information is mapped to a plurality of first participles, respectively extract a plurality of preset index tables corresponding to the plurality of first participles; wherein each index table includes the first The information of the video resource data to which the word segmentation belongs, and all the word segmentation in the video resource data; all the word segmentation in the video resource data is by grabbing the video resource data, extracting the feature text information of the video resource data, for The feature text information is generated by word segmentation;

In the specific implementation, the search engine can be used in advance to grab the video resource data on each platform through the crawler, and then build an index and build a database: extract the characteristic text information of the video resource data for word segmentation processing, and establish an index table corresponding to each word segment, The index table can store the information of video resource data (it can be ID, internal network address, external network address, etc. video identifiers, or a record composed of current word segmentation and other word segmentation), all information in video resource data participle (including the first participle and the second participle except the first participle).

In a preferred embodiment of the present invention, the feature text information may include video title, video keywords and/or video description.

Sub-step S52, using the plurality of index tables to determine the main participle, the main participle being the first participle corresponding to the index table with the largest total information of video resource data;

In order to improve user experience, for multiple first participles with relatively large differences in video resource data, the first participle with a small amount of video resource data information may be ignored. For example, for the first participle "Mid-Autumn Festival" and "Mooncake" mapped to the feature text information "Mid-Autumn Festival Mooncake", the total number of video resource data information in the index table of "Mid-Autumn Festival" is 900, while in the index table of "Mooncake" If the total number of video resource data information in the table is 100, "Mid-Autumn Festival" can be set as the main participle.

Sub-step S53, calculating the co-occurrence rate of each second participle in the main participle and its corresponding index table, the co-occurrence rate is the same as the number of occurrences of each second participle in the index table and the video resources in the index table The ratio of the total number of data information; wherein, the second participle is a participle except the first participle among all participles in the video resource data;

In the embodiment of the present invention, the co-occurrence rate of the principal participle may be used as the final co-occurrence rate.

Sub-step S54, extracting the second participle whose co-occurrence rate is higher than the preset threshold as the associated second participle.

In a specific implementation, the preset threshold may be set by those skilled in the art according to actual conditions, which is not limited in this embodiment of the present invention. The associated second participle extracted in the embodiment of the present invention may be empty, or may be one or more.

Step 104, obtaining the network link address of the second video resource data matching the one or more first participle and the associated second participle;

Specifically, after substep S33, a combination of the current first participle and one or more associated second participle can be obtained. For example, the feature text information is "dota", and the words with a high co-occurrence rate are: "funny", "egg pain", "2009", "Haitao", "first perspective" and "classic", co-occurrence The ratios are 40%, 35%, 30%, 25%, 20% and 10% respectively, then the obtained combinations are "dota funny", "dota egg pain", "dota2009", "dota Haitao", "dota First View" and "dota classic".

After sub-step S46, a combination of the current multiple first participle and one or more associated second participle may be obtained. For example, the feature text information is "Square Dance Binggege", map it to the first participle "Square Dance" and "Binggege", and extract the second participle that appears at the same time as these two first participle, such as the second participle " teaching", which can be used as the second participle of the association, and finally obtain the combination "big brother teaching square dancing soldier".

In a preferred embodiment of the present invention, step 104 may specifically include the following sub-steps:

Sub-step S61, acquiring the network link addresses of the main participle and the second video resource data associated with the second participle.

After sub-step S54, a combination of the current main participle and one or more associated second participle may be obtained. For example, for the first participle "Mid-Autumn Festival" and "Mooncake" mapped to the feature text information "Mid-Autumn Festival Mooncake", you can set "Mid-Autumn" as the main participle, get the associated second participle "Moon", and finally get the combination "Mid-Autumn Festival moon".

In the embodiment of the present invention, the search for matching video data resources can be performed based on a combination of one or more first participle and second participle, and when found, record its network connection address, which can specifically be an intranet address, or is the external network address.

Step 105, pushing the network link address of the second video resource data.

In practical applications, the network link address of the second video resource data can be placed at any position on the current page, or can be pushed by embedding icons or buttons, etc., and the user can trigger the network link address of the second video resource data to load The video data resource.

The present invention can push according to the existing released content, so that the search engine can get rid of the dependence on the user's search habit, and push out the video resource data that although few users search, but the video library has more related resources, so as to realize In-depth excavation of high-quality resources in the video library improves the efficiency of resource mining; in addition, the index table will continue to expand with the continuous accumulation of Internet video content, and the quantity and breadth of content produced by major video stations will far exceed that of users. The number of searched words is conducive to expanding the recall rate.

In the present invention, by obtaining the network connection address of the second video resource data matched by the first participle and the second participle, the user can directly obtain the video data resource based on this address, so that the user can obtain more results by simply searching without The search is submitted multiple times, thereby reducing the burden of accessing the server, reducing the occupation of network resources, and improving user experience.

For the method embodiment, for the sake of simple description, it is expressed as a series of action combinations, but those skilled in the art should know that the embodiment of the present invention is not limited by the described action order, because according to the embodiment of the present invention , certain steps may be performed in other order or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification belong to preferred embodiments, and the actions involved are not necessarily required by the embodiments of the present invention.

Referring to FIG. 2 , it shows a structural block diagram of an embodiment of an apparatus for pushing associated resource addresses based on video search according to an embodiment of the present invention, which may specifically include the following modules:

The characteristic text information acquisition module 201 is adapted to obtain the characteristic text information of the first video resource data when receiving a loading or playing request of the first video resource data;

The first participle mapping module 202 is adapted to map the feature text information into one or more first participles;

The second participle search module 203 is adapted to search for the associated second participle whose co-occurrence rate with the one or more first participle is higher than the preset threshold; the co-occurrence rate of the current one or more first participle and The probability that the second participle co-occurs in the same video resource data;

The network connection address obtaining module 204 is adapted to obtain the network link address of the second video resource data matching the one or more first participle and the associated second participle;

The network connection address pushing module 205 is adapted to push the network link address of the second video resource data.

In a preferred embodiment of the present invention, the feature text information acquisition module can also be adapted to:

When receiving the play request of the first video data, receiving the characteristic text information of the first video resource data sent by the current terminal;

or,

When the first video data loading request is received, the feature text information of the video resource data preset locally is extracted.

In a preferred embodiment of the present invention, the first word segmentation mapping module can also be adapted to:

Extracting a participle mapped by the text information of the feature book;

or,

When the received feature text information is a compound word, split the feature text information into multiple search subwords; and extract multiple word segmentations mapped to the multiple search subwords.

In a preferred embodiment of the present invention, the second word segmentation search module can also be adapted to:

When the feature text information is mapped to a first participle, extract a preset index table corresponding to the first participle; wherein, the index table includes information about the video resource data to which the first participle belongs, and , all the word segmentations in the video resource data; all the word segmentations in the video resource data are generated by capturing the video resource data, extracting the feature text information of the video resource data, and performing word segmentation on the feature text information;

Calculate the co-occurrence rate of the first participle and each second participle in the index table, the co-occurrence rate is the number of occurrences of each second participle in the index table and the information of the video resource data in the index table The ratio of the total number; wherein, the second participle is a participle except the first participle among all participles in the video resource data;

Extracting the second participle whose co-occurrence rate is higher than the preset threshold as the associated second participle.

In a preferred embodiment of the present invention, the second word segmentation search module can also be adapted to:

When the feature text information is mapped to a plurality of first participles, a plurality of preset index tables corresponding to the plurality of first participles are respectively extracted; each index table includes the video resource data to which the first participle belongs information, and all the word segmentation in the video resource data; all the word segmentation in the video resource data is to extract the feature text information of the video resource data by grabbing the video resource data, and carry out the feature text information on the feature text information participle generation;

Extracting a second word segment that co-occurs with the plurality of first word segments as a candidate word segment; wherein, the second word segment is a word segment in all word segments in the video resource data except the first word segment;

Calculate the co-occurrence rate of the first participle and the candidate participle in each index table respectively, the co-occurrence rate is the number of occurrences of the candidate participle in the index table and the total information of the video resource data in the index table ratio of

Configure a plurality of weights corresponding to the co-occurrence ratios of the plurality of first participle and the candidate participle respectively;

Calculating the average of multiple weighted co-occurrence rates, respectively, as the co-occurrence rates of the multiple first participle and the candidate participle;

Extracting the candidate participle whose co-occurrence rate is higher than the preset threshold as the associated second participle.

In a preferred embodiment of the present invention, the second word segmentation search module can also be adapted to:

When the feature text information is mapped to a plurality of first participles, a plurality of preset index tables corresponding to the plurality of first participles are respectively extracted; wherein, each index table includes the video to which the first participle belongs The information of the resource data, and all word segmentations in the video resource data; all the word segmentations in the video resource data are to extract the feature text information of the video resource data by grabbing the video resource data, and to the feature text information for word segmentation;

Using the plurality of index tables to determine the main participle, the main participle is the first participle corresponding to the index table with the largest total information of the video resource data;

Calculate the co-occurrence rate of each second participle in the main participle and its corresponding index table, the co-occurrence rate is the number of occurrences of each second participle in the index table and the total number of video resource data information in the index table ratio; wherein, the second participle is a participle except the first participle among all participles in the video resource data;

Extracting the second participle whose co-occurrence rate is higher than the preset threshold as the associated second participle.

In a preferred embodiment of the present invention, the feature text information may include video title, video keywords and/or video description.

In a preferred embodiment of the present invention, the network connection address obtaining module can also be adapted to:

Obtain the network link addresses of the main participle and the second video resource data associated with the second participle.

As for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and for related parts, please refer to the part of the description of the method embodiment.

The algorithms and displays presented herein are not inherently related to any particular computer, virtual system, or other device. Various generic systems can also be used with the teachings based on this. The structure required to construct such a system is apparent from the above description. Furthermore, the present invention is not specific to any particular programming language. It should be understood that various programming languages can be used to implement the content of the present invention described herein, and the above description of specific languages is for disclosing the best mode of the present invention.

In the description provided herein, numerous specific details are set forth. However, it is understood that embodiments of the invention may be practiced without these specific details. In some instances, well-known methods, structures and techniques have not been shown in detail in order not to obscure the understanding of this description.

Similarly, it should be appreciated that in the foregoing description of exemplary embodiments of the invention, in order to streamline this disclosure and to facilitate an understanding of one or more of the various inventive aspects, various features of the invention are sometimes grouped together in a single embodiment, figure, or its description. This method of disclosure, however, is not to be interpreted as reflecting an intention that the claimed invention requires more features than are expressly recited in each claim. Rather, as the following claims reflect, inventive aspects lie in less than all features of a single foregoing disclosed embodiment. Thus, the claims following the Detailed Description are hereby expressly incorporated into this Detailed Description, with each claim standing on its own as a separate embodiment of this invention.

Those skilled in the art can understand that the modules in the device in the embodiment can be adaptively changed and arranged in one or more devices different from the embodiment. Modules or units or components in the embodiments may be combined into one module or unit or component, and furthermore may be divided into a plurality of sub-modules or sub-units or sub-assemblies. All features disclosed in this specification (including accompanying claims, abstract and drawings), as well as any method or method so disclosed, may be used in any combination, except that at least some of such features and/or processes or units are mutually exclusive. All processes or units of equipment are combined. Each feature disclosed in this specification (including accompanying claims, abstract and drawings) may be replaced by alternative features serving the same, equivalent or similar purpose, unless expressly stated otherwise.

Furthermore, those skilled in the art will understand that although some embodiments described herein include some features included in other embodiments but not others, combinations of features from different embodiments are meant to be within the scope of the invention. and form different embodiments. For example, in the following claims, any of the claimed embodiments may be used in any combination.

The various component embodiments of the present invention may be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. Those skilled in the art should understand that a microprocessor or a digital signal processor (DSP) may be used in practice to implement some or all of the components in the video search-based associated resource address push device according to the embodiment of the present invention Or full functionality. The present invention can also be implemented as an apparatus or an apparatus program (for example, a computer program and a computer program product) for performing a part or all of the methods described herein. Such a program for realizing the present invention may be stored on a computer-readable medium, or may be in the form of one or more signals. Such a signal may be downloaded from an Internet site, or provided on a carrier signal, or provided in any other form.

It should be noted that the above-mentioned embodiments illustrate rather than limit the invention, and that those skilled in the art will be able to design alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word "comprising" does not exclude the presence of elements or steps not listed in a claim. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The invention can be implemented by means of hardware comprising several distinct elements, and by means of a suitably programmed computer. In a unit claim enumerating several means, several of these means can be embodied by one and the same item of hardware. The use of the words first, second, and third, etc. does not indicate any order. These words can be interpreted as names.

The invention discloses A1, a method for pushing associated resource addresses based on video search, including:

When receiving the loading or playing request of the first video resource data, acquire the feature text information of the first video resource data;

Mapping the feature text information into one or more first participles;

Finding the associated second participle whose co-occurrence rate with the one or more first participle is higher than the preset threshold; the co-occurrence rate of the current one or more first participle and the second participle are in the same video resource data probability of co-occurrence;

Acquire the network link address of the second video resource data matching the one or more first participle and the associated second participle;

Pushing the network link address of the second video resource data.

A2. The method as described in A1, when the loading or playing request of the first video resource data is received, the step of obtaining the feature text information of the first video resource data includes:

When receiving the play request of the first video data, receiving the characteristic text information of the first video resource data sent by the current terminal;

or,

When the first video data loading request is received, the feature text information of the video resource data preset locally is extracted.

A3. The method as described in A1, the step of mapping the feature text information into one or more first word segmentations includes:

Extracting a participle mapped by the text information of the feature book;

or,

When the received feature text information is a compound word, split the feature text information into multiple search subwords; and extract multiple word segmentations mapped to the multiple search subwords.

A4, the method as described in A1, the step of searching for the associated second participle whose co-occurrence rate with the one or more first participle is higher than the preset threshold includes:

When the feature text information is mapped to a first participle, extract a preset index table corresponding to the first participle; wherein, the index table includes information about the video resource data to which the first participle belongs, and , all the word segmentations in the video resource data; all the word segmentations in the video resource data are generated by capturing the video resource data, extracting the feature text information of the video resource data, and performing word segmentation on the feature text information;

Calculate the co-occurrence rate of the first participle and each second participle in the index table, the co-occurrence rate is the number of occurrences of each second participle in the index table and the information of the video resource data in the index table The ratio of the total number; wherein, the second participle is a participle except the first participle among all participles in the video resource data;

Extracting the second participle whose co-occurrence rate is higher than the preset threshold as the associated second participle.

A5, the method as described in A1, the step of searching for the associated second participle whose co-occurrence rate with the one or more first participle is higher than the preset threshold includes:

When the feature text information is mapped to a plurality of first participles, a plurality of preset index tables corresponding to the plurality of first participles are respectively extracted; each index table includes the video resource data to which the first participle belongs information, and all the word segmentation in the video resource data; all the word segmentation in the video resource data is to extract the feature text information of the video resource data by grabbing the video resource data, and carry out the feature text information on the feature text information participle generation;

Extracting a second word segment that co-occurs with the plurality of first word segments as a candidate word segment; wherein, the second word segment is a word segment in all word segments in the video resource data except the first word segment;

Calculate the co-occurrence rate of the first participle and the candidate participle in each index table respectively, the co-occurrence rate is the number of occurrences of the candidate participle in the index table and the total information of the video resource data in the index table ratio of

Configure a plurality of weights corresponding to the co-occurrence ratios of the plurality of first participle and the candidate participle respectively;

Calculating the average of multiple weighted co-occurrence rates, respectively, as the co-occurrence rates of the multiple first participle and the candidate participle;

Extracting the candidate participle whose co-occurrence rate is higher than the preset threshold as the associated second participle.

A6, the method as described in A1, the step of searching for the associated second participle whose co-occurrence rate with the one or more first participle is higher than the preset threshold includes:

When the feature text information is mapped to a plurality of first participles, a plurality of preset index tables corresponding to the plurality of first participles are respectively extracted; wherein, each index table includes the video to which the first participle belongs The information of the resource data, and all word segmentations in the video resource data; all the word segmentations in the video resource data are to extract the feature text information of the video resource data by grabbing the video resource data, and to the feature text information for word segmentation;

Using the plurality of index tables to determine the main participle, the main participle is the first participle corresponding to the index table with the largest total information of the video resource data;

Calculate the co-occurrence rate of each second participle in the main participle and its corresponding index table, the co-occurrence rate is the number of occurrences of each second participle in the index table and the total number of video resource data information in the index table ratio; wherein, the second participle is a participle except the first participle among all participles in the video resource data;

Extracting the second participle whose co-occurrence rate is higher than the preset threshold as the associated second participle.

A7. The method as described in A1 or A4 or A5 or A6, wherein the feature text information includes video title, video keywords and/or video description.

A8. The method as described in A6, the step of obtaining the network link address of the second video resource data matching the one or more first participle and the associated second participle includes:

Obtaining the network link addresses of the main participle and the second video resource data associated with the second participle.

The present invention also discloses B9, a device for pushing associated resource addresses based on video search, including:

The characteristic text information acquisition module is adapted to obtain the characteristic text information of the first video resource data when receiving the loading or playing request of the first video resource data;

The first participle mapping module is adapted to map the feature text information into one or more first participles;

The second participle search module is adapted to search for the associated second participle whose co-occurrence rate with the one or more first participle is higher than the preset threshold; the co-occurrence rate of the current one or more first participle and the second The probability that dichotomous words co-occur in the same video resource data;

A network connection address obtaining module, adapted to obtain the network link address of the second video resource data matching the one or more first participle and the associated second participle;

The network connection address pushing module is adapted to push the network link address of the second video resource data.

B10, the device as described in B9, the feature text information acquisition module is also suitable for:

When receiving the play request of the first video data, receiving the characteristic text information of the first video resource data sent by the current terminal;

or,

When the first video data loading request is received, the feature text information of the video resource data preset locally is extracted.

B11, the device as described in B9, the first word segmentation mapping module is also suitable for:

Extracting a participle mapped by the text information of the feature book;

or,

When the received feature text information is a compound word, split the feature text information into multiple search subwords; and extract multiple word segmentations mapped to the multiple search subwords.

B12, the device as described in B9, the second participle search module is also suitable for:

When the feature text information is mapped to a first participle, extract a preset index table corresponding to the first participle; wherein, the index table includes information about the video resource data to which the first participle belongs, and , all the word segmentations in the video resource data; all the word segmentations in the video resource data are generated by capturing the video resource data, extracting the feature text information of the video resource data, and performing word segmentation on the feature text information;

Calculate the co-occurrence rate of the first participle and each second participle in the index table, the co-occurrence rate is the number of occurrences of each second participle in the index table and the information of the video resource data in the index table The ratio of the total number; wherein, the second participle is a participle except the first participle among all participles in the video resource data;

Extracting the second participle whose co-occurrence rate is higher than the preset threshold as the associated second participle.

B13, the device as described in B9, described second participle search module is also suitable for:

When the feature text information is mapped to a plurality of first participles, a plurality of preset index tables corresponding to the plurality of first participles are respectively extracted; each index table includes the video resource data to which the first participle belongs information, and all the word segmentation in the video resource data; all the word segmentation in the video resource data is to extract the feature text information of the video resource data by grabbing the video resource data, and carry out the feature text information on the feature text information participle generation;

Extracting a second word segment that co-occurs with the plurality of first word segments as a candidate word segment; wherein, the second word segment is a word segment in all word segments in the video resource data except the first word segment;

Calculate the co-occurrence rate of the first participle and the candidate participle in each index table respectively, the co-occurrence rate is the number of occurrences of the candidate participle in the index table and the total information of the video resource data in the index table ratio of

Configure a plurality of weights corresponding to the co-occurrence ratios of the plurality of first participle and the candidate participle respectively;

Calculating the average of multiple weighted co-occurrence rates, respectively, as the co-occurrence rates of the multiple first participle and the candidate participle;

Extracting the candidate participle whose co-occurrence rate is higher than the preset threshold as the associated second participle.

B14, the device as described in B9, described second participle search module is also suitable for:

When the feature text information is mapped to a plurality of first participles, a plurality of preset index tables corresponding to the plurality of first participles are respectively extracted; wherein, each index table includes the video to which the first participle belongs The information of the resource data, and all word segmentations in the video resource data; all the word segmentations in the video resource data are to extract the feature text information of the video resource data by grabbing the video resource data, and to the feature text information for word segmentation;

Using the plurality of index tables to determine the main participle, the main participle is the first participle corresponding to the index table with the largest total information of the video resource data;

Calculate the co-occurrence rate of each second participle in the main participle and its corresponding index table, the co-occurrence rate is the number of occurrences of each second participle in the index table and the total number of video resource data information in the index table ratio; wherein, the second participle is a participle except the first participle among all participles in the video resource data;

Extracting the second participle whose co-occurrence rate is higher than the preset threshold as the associated second participle.

B15. The device as described in B9 or B12 or B13 or B14, wherein the feature text information includes video title, video keywords and/or video description.

B16, the device as described in B14, the network connection address acquisition module is also suitable for:

Obtain the network link addresses of the main participle and the second video resource data associated with the second participle.

Claims (10)

1. A method for pushing associated resource addresses based on video search, comprising: When receiving the loading or playing request of the first video resource data, acquire the feature text information of the first video resource data; Mapping the feature text information into one or more first participles; Finding the associated second participle whose co-occurrence rate with the one or more first participle is higher than the preset threshold; the co-occurrence rate of the current one or more first participle and the second participle are in the same video resource data probability of co-occurrence; Acquire the network link address of the second video resource data matching the one or more first participle and the associated second participle; Pushing the network link address of the second video resource data. 2. The method according to claim 1, wherein the step of obtaining the feature text information of the first video resource data comprises: When receiving the play request of the first video data, receiving the characteristic text information of the first video resource data sent by the current terminal; or, When the first video data loading request is received, the feature text information of the video resource data preset locally is extracted. 3. The method according to claim 1, wherein the step of mapping the feature text information into one or more first word segmentations comprises: Extracting a participle mapped by the text information of the feature book; or, When the received feature text information is a compound word, split the feature text information into multiple search subwords; and extract multiple word segmentations mapped to the multiple search subwords. 4. The method according to claim 1, wherein the step of searching for the associated second participle whose co-occurrence rate with the one or more first participle is higher than a preset threshold comprises: When the feature text information is mapped to a first participle, extract a preset index table corresponding to the first participle; wherein, the index table includes information about the video resource data to which the first participle belongs, and , all the word segmentations in the video resource data; all the word segmentations in the video resource data are generated by capturing the video resource data, extracting the feature text information of the video resource data, and performing word segmentation on the feature text information; Calculate the co-occurrence rate of the first participle and each second participle in the index table, the co-occurrence rate is the number of occurrences of each second participle in the index table and the information of the video resource data in the index table The ratio of the total number; wherein, the second participle is a participle except the first participle among all participles in the video resource data; Extracting the second participle whose co-occurrence rate is higher than the preset threshold as the associated second participle. 5. The method according to claim 1, wherein the step of searching for the associated second participle whose co-occurrence rate with the one or more first participle is higher than a preset threshold comprises: When the feature text information is mapped to a plurality of first participles, a plurality of preset index tables corresponding to the plurality of first participles are respectively extracted; each index table includes the video resource data to which the first participle belongs information, and all the word segmentation in the video resource data; all the word segmentation in the video resource data is to extract the feature text information of the video resource data by grabbing the video resource data, and carry out the feature text information on the feature text information participle generation; Extracting a second word segment that co-occurs with the plurality of first word segments as a candidate word segment; wherein, the second word segment is a word segment in all word segments in the video resource data except the first word segment; Calculate the co-occurrence rate of the first participle and the candidate participle in each index table respectively, the co-occurrence rate is the number of occurrences of the candidate participle in the index table and the total information of the video resource data in the index table ratio of Configure a plurality of weights corresponding to the co-occurrence ratios of the plurality of first participle and the candidate participle respectively; Calculating the average of multiple weighted co-occurrence rates, respectively, as the co-occurrence rates of the multiple first participle and the candidate participle; Extracting the candidate participle whose co-occurrence rate is higher than the preset threshold as the associated second participle. 6. The method according to claim 1, wherein the step of searching for an associated second participle whose co-occurrence rate with the one or more first participle is higher than a preset threshold comprises: When the feature text information is mapped to a plurality of first participles, a plurality of preset index tables corresponding to the plurality of first participles are respectively extracted; wherein, each index table includes the video to which the first participle belongs The information of the resource data, and all word segmentations in the video resource data; all the word segmentations in the video resource data are to extract the feature text information of the video resource data by grabbing the video resource data, and to the feature text information for word segmentation; Using the plurality of index tables to determine the main participle, the main participle is the first participle corresponding to the index table with the largest total information of the video resource data; Calculate the co-occurrence rate of each second participle in the main participle and its corresponding index table, the co-occurrence rate is the number of occurrences of each second participle in the index table and the total number of video resource data information in the index table ratio; wherein, the second participle is a participle except the first participle among all participles in the video resource data; Extracting the second participle whose co-occurrence rate is higher than the preset threshold as the associated second participle. 7. The method according to claim 1 or 4 or 5 or 6, wherein the feature text information includes video title, video keywords and/or video description. 8. The method according to claim 6, wherein the step of obtaining the network link address of the second video resource data matching the one or more first participle and the associated second participle comprises: Obtaining the network link addresses of the main participle and the second video resource data associated with the second participle. 9. A device for pushing associated resource addresses based on video search, comprising: The characteristic text information acquisition module is adapted to obtain the characteristic text information of the first video resource data when receiving the loading or playing request of the first video resource data; The first participle mapping module is adapted to map the feature text information into one or more first participles; The second participle search module is adapted to search for the associated second participle whose co-occurrence rate with the one or more first participle is higher than the preset threshold; the co-occurrence rate of the current one or more first participle and the second The probability that the dichotomous words co-occur in the same video resource data; A network connection address obtaining module, adapted to obtain the network link address of the second video resource data matching the one or more first participle and the associated second participle; The network connection address pushing module is adapted to push the network link address of the second video resource data. 10. The device according to claim 9, wherein the characteristic text information acquisition module is also suitable for: When receiving the play request of the first video data, receiving the characteristic text information of the first video resource data sent by the current terminal; or, When the first video data loading request is received, the feature text information of the video resource data preset locally is extracted.

Images