WO2018227908A1 - Advertising keyword recommendation method and device, and advertising method and device - Google Patents

Abstract

An advertising keyword recommendation method and device, and an advertising method and device. The advertising keyword recommendation method comprises: acquiring brand words of a brand (S10); acquiring search information, mining concern words concerning the brand in the search information according to the brand words, and marking concern points of the concern words (S30); and recommending any concern word set corresponding to the brand words and the concern points as an advertising keyword (S50). According to the method and device, by mining the concern words concerning the brand in the search information by use of the brand words, and marking the concern points of the concern words, a plurality of concern word sets with a high concern on the brand and with the concern points being units are acquired, and finally any concern word set corresponding to the brand words and the concern points is recommended as the advertising keyword. The serving coverage is ensured by means of the synonymous concern words corresponding to the concern points while serving precision is ensured, and therefore the advertising effect is effectively ensured.

Description

Advertising keyword recommendation method and device, advertisement delivery method and device

Cross-reference to related applications

This application claims Baidu Online Network Technology (Beijing) Co., Ltd. submitted on June 12, 2017, the invention name is "Advertising keyword recommendation method and device, advertising method and device", Chinese patent application number "201710439477.3" Priority.

Technical field

The present invention relates to the field of advertisement delivery technologies, and in particular, to an advertisement placement keyword recommendation method and device, an advertisement delivery method and device.

Background technique

Search Engine Marketing (SEM) is a very effective online marketing channel that promotes a variety of business information, such as brand advertising. In search engine marketing, a customer purchases a keyword, and if the keyword is included in the content searched by the netizen, the brand advertisement of the customer is triggered.

Current solutions require customer-defined keywords or use system-recommended keywords. The shortcoming of the former is that custom keywords are only customized by experience or intuition. Since the netizens are not aware of the brand's concerns, custom keywords are often lacking in relevance; while the latter's drawback is that the system usually only recommends that customers have purchased the key. The related words of the word are less relevant to the brand. Therefore, the defects of the existing schemes may result in the quality of the keywords being difficult to guarantee the effect of the advertisements, which may result in poor advertising effects.

Summary of the invention

In view of the above-mentioned defects or deficiencies in the prior art, it is desirable to provide an advertisement placement keyword recommendation method and apparatus for identifying an advertisement delivery effect by identifying a network owner's attention to a brand, and an advertisement delivery method and apparatus.

In a first aspect, the present invention provides a method for recommending an advertisement placement keyword, including:

Get the brand word of the brand;

Obtaining search information, mining the attention words of the brand in the search information according to the brand words, and marking the attention points of the attention words;

The combination of the brand word and any of the attention words corresponding to the point of interest is recommended as the keyword for the advertisement.

In a second aspect, the present invention provides an advertisement delivery method, which comprises: using a combination of a brand word recommended by the advertisement keyword recommendation method and a focus word corresponding to a focus point as a keyword to perform advertisement placement.

In a third aspect, the present invention provides an advertisement placement keyword recommendation apparatus, including a brand word acquisition unit, a focus point mining unit, and a keyword recommendation unit.

Wherein, the brand word acquisition unit is configured to acquire a brand word of the brand;

The point of interest mining unit is configured to acquire search information, and mine the attention words of the brand in the search information according to the brand words, and mark the attention points of the attention words;

The keyword recommendation unit is configured to recommend a combination of a brand word and any attention word corresponding to the point of interest as a keyword for advertisement placement.

In a fourth aspect, the present invention provides an advertisement delivery device, including the above advertisement placement keyword recommendation device, and a delivery unit.

The delivery unit is configured to perform advertisement placement by using a combination of the recommended brand words and any attention words corresponding to the attention points as keywords.

In a fifth aspect, the present invention also provides an apparatus comprising one or more processors and a memory, wherein the memory includes instructions executable by the one or more processors to cause the one or more processors to perform each of the The advertisement delivery keyword recommendation method or the advertisement delivery method provided by the embodiment.

In a sixth aspect, the present invention further provides a computer readable storage medium storing a computer program, the computer program causing a computer to execute an advertisement placement keyword recommendation method or an advertisement delivery method according to various embodiments of the present invention.

The advertisement delivery keyword recommendation method and device provided by the embodiments of the present invention, the advertisement delivery method and the device, by using the brand word to mine the search information, pay attention to the attention word of the brand, and mark the attention point of the attention word, and obtain a plurality of A high-attention collection of attention words in focus units, and finally recommending a combination of a brand word and a focus word corresponding to a point of interest as a keyword for advertising, while ensuring accuracy of delivery, through attention point The synonymous or synonymous attention word guarantees the coverage of the delivery, thus achieving the effect of effectively ensuring the advertisement delivery;

The advertisement delivery keyword recommendation method and device, the advertisement delivery method and device provided by some embodiments of the present invention further realize automatic and accurate identification of brand words in the company name, thereby eliminating the need for the user to manually set the brand words.

DRAWINGS

Other features, objects, and advantages of the present application will become more apparent from the detailed description of the accompanying drawings.

FIG. 1 is a flowchart of a method for recommending an advertisement placement keyword according to an embodiment of the present invention.

2 is a flow chart of step S30 in a preferred embodiment of the method of FIG. 1.

3 is a flow chart of a preferred embodiment of the method of FIG. 1.

4 is a flow chart of step S11 in a preferred embodiment of the method of FIG.

Figure 5 is a flow diagram of a preferred embodiment of the method of Figure 4.

6 is a flow chart of a preferred embodiment of the method of FIG.

FIG. 7 is a flowchart of an advertisement delivery method according to an embodiment of the present invention.

FIG. 8 is a schematic structural diagram of an advertisement placement keyword recommendation apparatus according to an embodiment of the present invention.

Figure 9 is a schematic view showing the structure of a preferred embodiment of the apparatus shown in Figure 8.

Figure 10 is a schematic view showing the structure of a preferred embodiment of the apparatus shown in Figure 8.

Figure 11 is a schematic view showing the structure of a preferred embodiment of the apparatus shown in Figure 10.

Figure 12 is a schematic view showing the structure of a preferred embodiment of the apparatus shown in Figure 11.

FIG. 13 is a schematic structural diagram of an advertisement placing apparatus according to an embodiment of the present invention.

FIG. 14 is a schematic structural diagram of a device according to an embodiment of the present invention.

detailed description

The present application will be further described in detail below with reference to the accompanying drawings and embodiments. It is understood that the specific embodiments described herein are merely illustrative of the invention, rather than the invention. It should also be noted that, for the convenience of description, only parts related to the invention are shown in the drawings.

It should be noted that the embodiments in the present application and the features in the embodiments may be combined with each other without conflict. The present application will be described in detail below with reference to the accompanying drawings.

FIG. 1 is a flowchart of a method for recommending an advertisement placement keyword according to an embodiment of the present invention.

As shown in FIG. 1 , in the embodiment, the advertisement placement keyword recommendation method provided by the present invention includes:

S10: Obtain the brand word of the brand;

S30: acquiring search information, exchanging the attention words of the brand in the search information according to the brand words, and marking the attention points of the attention words;

S50: Recommend the combination of the brand word and any attention word corresponding to the point of interest as the keyword of the advertisement placement.

In step S10, the brand word is obtained in the following two ways in the embodiment: identifying the brand word in the company name of the brand, and acquiring the brand word of the brand set by the user. Among them, the brand word refers to the word that constitutes the font size in the company name, and the brand word may be the same as the brand name, or different but has a strong correlation with the brand.

Specifically, when the company name includes the brand name, the recognized brand word includes the brand name, and the user can further supplement the setting of the nickname and abbreviation of the current brand, and the names, nicknames, abbreviations, etc. of other brands owned by the enterprise cannot pass. Identify the brand word obtained by the company name;

When the company name does not include the brand name, the recognized brand words still have a strong correlation with the brand. The netizens who can recognize the brand words usually know the relationship between the brand words and the brand, and the user can further supplement the design. The name, nickname, abbreviation of the current brand, and the brand name, nickname, abbreviation, etc. of other brands owned by the company cannot be obtained by identifying the company name.

2 is a flow chart of step S30 in a preferred embodiment of the method of FIG. 1. As shown in FIG. 2, in this embodiment, step S30 includes:

S31: Acquire a plurality of search information, and filter search information that does not include a brand word;

S33: cutting the filtered search information, removing the brand words, and sorting the frequency of the words to select a number of high frequency attention words;

S35: Label the concerns of each word of interest.

Specifically, in step S31, the search information of a large number of netizens is acquired through the background of the search engine, and the unrelated search information is filtered by using the brand word obtained in step S10, thereby filtering out the search information that concerns the brand;

In step S33, each search information selected in step S31 is subjected to word-cutting, and words such as brand words and stop words are removed, and a plurality of sequences of attention words are obtained. Among them, the stop words specifically refer to words such as modal particles that do not reflect the attention of netizens to the brand. Counting the word frequency of each word in a sequence of attention words and performing reverse ordering, thereby selecting a number of words with the highest frequency of words;

In step S35, the points of interest of the attention words selected in step S33 are marked. For example, the attention points of the attention words such as "price reduction", "promotion", and "discount" are "price", "feel", "comfort" The focus of attention such as "durable" is "performance", and so on. Specifically, it can be marked by means of machine learning, or displayed for manual labeling.

Through the mining and labeling of step S30, several points of interest can be obtained, and each point of interest corresponds to a plurality of words with the same meaning, near meaning or high degree of association.

In a further embodiment, in the step S30, different mining algorithms commonly used in the field can also be used to mine the word of interest related to the brand word, and the same technical effect can be achieved.

In step S50, the combination of the recommended brand word and any of the attention words corresponding to the point of interest, for example, the combination of the recommended brand word “Sheng Da” and the attention point “price” corresponds to any of the attention words: When the group's recommended keywords are used for advertising, as long as the search information includes the brand word “Shengda” and the “price reduction” corresponding to “price reduction/promotion/discount”, the advertisement will be triggered.

The above embodiment uses the brand words to mine the attention words of the brand in the search information, and points the attention points of the attention words, and obtains a plurality of attention words set with the attention point as the unit with high attention to the brand, and finally the brand words. The combination of any word of interest corresponding to the point of interest is recommended as the keyword for advertising, and the coverage of the delivery is guaranteed by the synonymous or synonymous attention words corresponding to the point of interest while ensuring the accuracy of the delivery. The effect of ad serving.

3 is a flow chart of a preferred embodiment of the method of FIG. 2. As shown in FIG. 3, step S10 includes at least one of steps S11 and S12, that is, in some preferred embodiments, step S10 may be configured to adopt either one of the above two modes or two according to actual needs. Ways to get brand words.

4 is a flow chart of step S11 in a preferred embodiment of the method of FIG. As shown in FIG. 4, in a preferred embodiment, step S11 specifically includes:

S115: cutting the name of the company to obtain a sequence of word segments;

S116: labeling the local nouns and the institutional words in the sequence of word segments;

S117: traversing the word segment sequence from the going to the end, obtaining a word segment subsequence between the first place noun and the first institution word;

S118: traversing the word segment subsequence from the back to the front, obtaining a delimiter word whose word frequency is less than the first threshold and the closeness of the next adjacent word is less than the second threshold; wherein the closeness is based on the word frequency of the two adjacent words and adjacent Phrase frequency determination;

S119: In the segmentation subsequence, the words and separators before the separator are marked as brand words, and the words after the separator are marked as industry words.

Specifically, in this embodiment, the closeness Aff(word ₁ , word ₂ ) of two adjacent words word ₁ and word ₂ is calculated as:

Among them, Freq(word ₁ ) is the word frequency of word ₁ , Freq(word ₂ ) is the word frequency of word ₂ , Freq _adj (word ₁ , word ₂ ) is the adjacent phrase frequency of word ₁ and word ₂ , and σ is the smoothing parameter. The above word frequency and adjacent phrase frequency are obtained by statistics, and the smoothing parameters are set and adjusted according to experience. In more embodiments, the tightness can also be configured as other different calculation methods, and the same technical effect can be achieved as long as the closeness of the two words is positively correlated with the adjacent phrase frequency.

For example, for the company name "Beijing Shengda Fire Protection Equipment Co., Ltd.", the above steps S115-S119 are performed in sequence:

In step S115, the word segmentation sequence "Beijing-Shengda-Fire-Equipment-Limited-Company" is obtained by cutting the word (and the "limited-company" is merged into "limited company" by amendment);

In step S116, the local noun "Beijing" and the institutional word "limited company" in the word segmentation sequence are marked;

In step S117, traversing the word segmentation sequence from the heading to obtain the word segmentation sequence "Shengda-Fireproof-Device" between the first place name "Beijing" and the first organization word "Company";

In step S118, the word segment sub-sequence is traversed from the back to the front, the word frequency of the "device" is greater than the first threshold, and the condition is not met; the word frequency of the "fire prevention" is less than the first threshold, but the "fire" and "device" are obtained by calculation. The tightness is greater than the second threshold, and the condition is not met; the word frequency of “Shengda” is less than the first threshold, and the closeness of “Shengda” and “Fireproof” is less than the second threshold, so “Shengda” is judged as a separator;

In step S119, the word "Shengda" in the word segmentation subsequence "Shengda-Fireproof Equipment" and its previous words are marked as brand words, and the "fire prevention" and "equipment" after "Shengda" are marked as industry. word.

Further, the above steps further include modifying the special type of company name, for example:

In the process of cutting the word in step S115, the method further includes modifying a special type of company name, for example, removing "(China)" in the company name, etc.;

In the traversal process of step S117, if the noun is not obtained, the word segment sequence before the first institution word is used as the word segment subsequence; and other correction means configured for different special types of company names.

The above embodiment further realizes automatic and accurate identification of brand words in the company name, thereby eliminating the need for the user to manually set the brand words.

Figure 5 is a flow diagram of a preferred embodiment of the method of Figure 4. As shown in FIG. 5, in a preferred embodiment, step S11 further includes:

S111: Obtain a number of company names and perform a word segmentation to obtain a sequence of certain word segments;

S112: Counting the word frequency of each word in a sequence of partial words, and the frequency of adjacent phrases of each pair of adjacent words;

S113: Calculate the closeness of each pair of adjacent words.

Specifically, in step S111, all the company names in the registration information table are obtained, and more company names (and deduplication) can be further obtained through other channels commonly used in the field, and each acquired company name is cut separately. Words to obtain a sequence of word segments;

In step S112, the word frequency of each word and the adjacent phrase frequency of each pair of adjacent words are obtained by statistics;

In step S113, the closeness of each pair of adjacent words is calculated according to the above formula (1). When the company name identified in step S11 is in the registration information table, step S118 simply calls the tightness calculated in advance in step S113 without further calculation at the time of judgment. In further embodiments, the tightness can also be calculated according to different calculations configured.

6 is a flow chart of a preferred embodiment of the method of FIG. As shown in FIG. 6, in a preferred embodiment, step S11 further includes:

S114: Mark high frequency words in each end word of several word segmentation sequences as institutional words.

Specifically, in this embodiment, the screening method of the high frequency words in step S114 is: the word frequency of the high frequency words is not less than the third threshold, and the ratio of the sum of the word frequencies of the high frequency words to the sum of the word frequencies of the last words Not less than the fourth threshold. In more embodiments, the same technical effects can be achieved by using machine training, manual marking, obtaining an existing institutional vocabulary for marking, and the like by different means.

FIG. 7 is a flowchart of an advertisement delivery method according to an embodiment of the present invention. As shown in FIG. 7 , an advertisement delivery method provided by this embodiment includes the advertisement delivery keyword recommendation method provided by any of the foregoing embodiments, and:

S70: Perform the advertisement placement by using the combination of the brand word recommended in step S50 and any attention word corresponding to the attention point as a keyword.

FIG. 8 is a schematic structural diagram of an advertisement placement keyword recommendation apparatus according to an embodiment of the present invention. The apparatus shown in Fig. 8 can correspondingly perform the method shown in Fig. 1.

As shown in FIG. 8, in the present embodiment, the present invention provides an advertisement placement keyword recommendation device 10, which includes a brand word acquisition unit 11, a focus point mining unit 13, and a keyword recommendation unit 15.

The brand word obtaining unit 11 is configured to acquire a brand word of the brand;

The point of interest mining unit 13 is configured to acquire search information, and mine the attention words of the brand in the search information according to the brand words, and mark the attention points of the attention words;

The keyword recommendation unit 15 is configured to recommend a combination of a brand word and any attention word corresponding to the point of interest as a keyword for advertisement placement.

For the specific recommendation principle, refer to the method shown in Figure 1, and details are not described here.

Figure 9 is a schematic view showing the structure of a preferred embodiment of the apparatus shown in Figure 8. The apparatus shown in FIG. 9 can correspondingly perform the method shown in FIG. 2.

As shown in FIG. 9, in a preferred embodiment, the point of interest mining unit 13 includes a filter subunit 131, a focus word mining subunit 133, and a point of interest labeling subunit 135.

The filtering sub-unit 131 is configured to acquire a plurality of search information, and filter the search information that does not include the brand word;

The attention word mining sub-unit 133 is configured to cut the filtered search information, remove the brand words and then sort the words to select a number of high-frequency attention words;

The attention point labeling sub-unit 135 is configured to mark the points of interest of the respective attention words.

Figure 10 is a schematic view showing the structure of a preferred embodiment of the apparatus shown in Figure 8. The apparatus shown in FIG. 10 can correspondingly perform the method shown in FIG.

As shown in FIG. 10, in a preferred embodiment, the brand word acquisition unit 11 includes at least one of the following:

a brand word recognition unit 111 configured to identify a brand word in a company name of the brand;

The brand word setting unit 113 is configured to acquire a brand word of the brand set by the user.

Figure 11 is a schematic view showing the structure of a preferred embodiment of the apparatus shown in Figure 10. The apparatus shown in Fig. 11 can correspondingly perform the method shown in Fig. 4.

As shown in FIG. 11, in a preferred embodiment, the brand word identifying unit 111 includes a pre-processing sub-unit 1113, a dividing sub-unit 1115, and a marking sub-unit 1117.

The pre-processing sub-unit 1113 is configured to cut a word for the company name, obtain a word segmentation sequence, mark the local nouns and the organization words in the word segmentation sequence, and traverse the word segment sequence from the arrival to obtain the first place noun and the first a segmentation sub-sequence between the institutional words (corresponding to steps S115-S117 of the method shown in Figure 4);

The dividing sub-unit 1115 is configured to traverse the word segment sub-sequence from the back to the front to obtain a delimiter word whose word frequency is smaller than the first threshold and whose closeness to the next adjacent word is less than the second threshold; wherein the closeness is based on two adjacent words Word frequency and adjacent phrase frequency determination (corresponding to step S118 of the method shown in Figure 4);

The tag subunit 1117 is configured to mark the word and the separator word before the separator word as the brand word in the word segment subsequence, and the word after the separator word as the industry word (corresponding to step S119 of the method shown in FIG. 4).

Figure 12 is a schematic view showing the structure of a preferred embodiment of the apparatus shown in Figure 11. The apparatus shown in Figure 12 can perform the method shown in Figures 5-6.

As shown in FIG. 12, in a preferred embodiment, the brand word recognition unit 111 further includes a data support subunit 1111. The data support sub-unit 1111 is configured to acquire a number of company names and perform word-cutting, obtain a sequence of partial words, count the word frequency of each word in the sequence of partial words, and the adjacent phrase frequency of each pair of adjacent words, and calculate each The closeness to adjacent words.

In a preferred embodiment, the brand word recognition unit 111 is further configured to mark high frequency words in each of the plurality of word segmentation sequences as institutional words.

FIG. 13 is a schematic structural diagram of an advertisement placing apparatus according to an embodiment of the present invention. The apparatus shown in Fig. 13 can correspondingly perform the method shown in Fig. 7.

As shown in FIG. 13 , the present invention further provides an advertisement placement device 20, which includes the advertisement placement keyword recommendation device 10 and the delivery unit 21 provided by any of the above embodiments.

The delivery unit 21 is configured to perform advertisement placement by using a combination of the brand word recommended by the keyword recommendation unit 15 and any of the attention words corresponding to the attention point as a keyword.

FIG. 14 is a schematic structural diagram of a device according to an embodiment of the present invention.

As shown in FIG. 14, as another aspect, the present application also provides an apparatus 1400 including one or more central processing units (CPUs) 1401, which may be according to programs stored in a read only memory (ROM) 1402 or Various appropriate actions and processes are performed from the program loaded into the random access memory (RAM) 1403 by the storage portion 1408. In the RAM 1403, various programs and data required for the operation of the device 1400 are also stored. The CPU 1401, the ROM 1402, and the RAM 1403 are connected to each other through a bus 1404. An input/output (I/O) interface 1405 is also coupled to bus 1404.

The following components are connected to the I/O interface 1405: an input portion 1406 including a keyboard, a mouse, etc.; an output portion 1407 including a cathode ray tube (CRT), a liquid crystal display (LCD), and the like, and a speaker; a storage portion 1408 including a hard disk or the like And a communication portion 1409 including a network interface card such as a LAN card, a modem, or the like. The communication section 1409 performs communication processing via a network such as the Internet. Driver 1410 is also coupled to I/O interface 1405 as needed. A removable medium 1411 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory or the like is mounted on the drive 1410 as needed so that a computer program read therefrom is installed into the storage portion 1408 as needed.

In particular, according to an embodiment of the present disclosure, the advertisement placement keyword recommendation method or the advertisement delivery method described in any of the above embodiments may be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product comprising a computer program tangibly embodied on a machine readable medium, the computer program comprising program code for performing an advertisement placement keyword recommendation method or an advertisement delivery method . In such an embodiment, the computer program can be downloaded and installed from the network via the communication portion 1409, and/or installed from the removable medium 1411.

In still another aspect, the present application further provides a computer readable storage medium, which may be a computer readable storage medium included in the apparatus of the above embodiment; or may exist separately, not assembled A computer readable storage medium in a device. The computer readable storage medium stores one or more programs that are used by one or more processors to perform an ad placement keyword recommendation method or an ad delivery method as described in this application.

The flowchart and block diagrams in the Figures illustrate the architecture, functionality and operation of possible implementations of systems, methods and computer program products in accordance with various embodiments of the invention. In this regard, each block of the flowchart or block diagram can represent a module, a program segment, or a portion of code that includes one or more of the logic functions for implementing the specified. Executable instructions. It should also be noted that in some alternative implementations, the functions noted in the blocks may also occur in a different order than that illustrated in the drawings. For example, two successively represented blocks may in fact be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending upon the functionality involved. It is also noted that each block of the block diagrams and/or flowcharts, and combinations of blocks in the block diagrams and/or flowcharts, can be implemented by a dedicated hardware-based system that performs the specified functions or operations. Or can be implemented by a combination of dedicated hardware and computer instructions.

The units or modules described in the embodiments of the present application may be implemented by software or by hardware. The described units or modules may also be provided in the processor. For example, each of the units may be a software program disposed in a computer or a mobile smart device, or may be a separately configured hardware device. The names of these units or modules do not in any way constitute a limitation on the unit or module itself.

The above description is only a preferred embodiment of the present application and a description of the principles of the applied technology. It should be understood by those skilled in the art that the scope of the invention referred to in the present application is not limited to the specific combination of the above technical features, and should also be covered by the above technical features or without departing from the concept of the present application. Other technical solutions formed by arbitrarily combining the equivalent features. For example, the above features are combined with the technical features disclosed in the present application, but are not limited to the technical features having similar functions.

Claims (20)

1. An advertisement placement keyword recommendation method, which is characterized by comprising:

   Get the brand word of the brand;

   Obtaining search information, excerpting the attention words of the brand in the search information according to the brand words, and marking the attention points of the attention words;

   The combination of the brand word and any attention word corresponding to the attention point is recommended as a keyword for advertisement placement.
2. The method according to claim 1, wherein the brand word of the acquired brand comprises at least one of the following:

   Identify brand words in the company name of the brand;

   Obtain the brand words of the brand set by the user.
3. The method of claim 2 wherein said identifying a brand word in a company name of said brand comprises:

   Cutting the name of the company to obtain a sequence of word segments;

   Labeling the nouns and institutional words in the sequence of word segments;

   Traversing the word segmentation sequence from the heading to obtain a word segmentation subsequence between the first land noun and the first institutional word;

   Traversing the word segment subsequence from the back to the front, obtaining a delimiter word whose word frequency is less than the first threshold and the closeness of the next adjacent word is less than the second threshold; wherein the tightness is based on the word frequency of two adjacent words and Adjacent phrase frequency is determined;

   In the word segmentation sequence, the word before the separator word and the separator word are marked as brand words, and the words after the separator word are marked as industry words.
4. The method according to claim 3, characterized in that the closeness Aff(word ₁ , word ₂ ) of two adjacent words word ₁ and word ₂ is calculated as:

   

   Among them, Freq(word ₁ ) is the word frequency of word ₁ , Freq(word ₂ ) is the word frequency of word ₂ , Freq _adj (word ₁ , word ₂ ) is the adjacent phrase frequency of word ₁ and word ₂ , and σ is the smoothing parameter.
5. The method according to claim 2 or 3, wherein the identifying the brand words in the company name of the brand further comprises:

   Obtain a number of company names and cut the words to obtain a sequence of several word segments;

   Counting the word frequency of each of the plurality of word segmentation sequences, and the adjacent phrase frequency of each pair of adjacent words;

   Calculate the closeness of each pair of adjacent words.
6. The method according to any one of claims 2 to 5, wherein the identifying the brand words in the company name of the brand further comprises:

   The high frequency words in the last words of the plurality of word segmentation sequences are marked as institutional words.
7. The method according to claim 6, wherein the screening method of the high frequency word is: the word frequency of the high frequency word is not less than the third threshold, and the sum of the word frequency of each high frequency word and the word frequency of each end word The ratio of the sum is not less than the fourth threshold.
8. The method according to any one of claims 1 to 7, wherein the acquiring the search information, mining the attention words of the brand in the search information according to the brand words, and marking the attention words Points of interest include:

   Obtaining a plurality of search information, filtering search information not including the brand word;

   Cutting the filtered search information, removing the brand words and sorting the word frequency and sorting to select a number of high frequency attention words;

   Label the concerns of each of the attention words.
9. An advertisement delivery method, characterized in that a combination of a brand word recommended by the advertisement placement keyword recommendation method according to any one of claims 1-8 and any attention word corresponding to a focus point is used as a keyword. Ad serving.
10. An advertisement placement keyword recommendation device, comprising:

    a brand word acquisition unit configured to obtain a brand word of the brand;

    a point of interest mining unit configured to acquire search information, and to mine the attention word of the brand in the search information according to the brand word, and mark the focus of the attention word;

    The keyword recommendation unit is configured to recommend the combination of the brand word and any attention word corresponding to the attention point as a keyword for advertisement placement.
11. The apparatus according to claim 10, wherein said brand word acquisition unit comprises at least one of the following:

    a brand word recognition unit configured to identify a brand word in a company name of the brand;

    The brand word setting unit is configured to acquire a brand word of the brand set by the user.
12. The device according to claim 11, wherein the brand word recognition unit comprises:

    a pre-processing sub-unit configured to cut a word for a company name, obtain a word segmentation sequence, mark a local noun and an organization word in the word segment sequence, and traverse the word segment sequence from the heading to obtain the first place noun and a segmentation subsequence between the first institutional words;

    Dividing a subunit, configured to traverse the word segment subsequence from back to front, obtaining a delimiter word whose word frequency is less than a first threshold and the closeness of the next adjacent word is less than a second threshold; wherein the tightness is according to two The word frequency of adjacent words and the frequency of adjacent phrases are determined;

    a mark subunit configured to mark a word before the separator word and the word separator as a brand word in the word segment subsequence, and the word after the separator word is marked as an industry word.
13. The apparatus according to claim 12, wherein the closeness Aff(word ₁ , word ₂ ) of two adjacent words word ₁ and word ₂ is calculated as:

    

    Among them, Freq(word ₁ ) is the word frequency of word ₁ , Freq(word ₂ ) is the word frequency of word ₂ , Freq _adj (word ₁ , word ₂ ) is the adjacent phrase frequency of word ₁ and word ₂ , and σ is the smoothing parameter.
14. The device according to claim 11 or 12, wherein the brand word recognition unit further comprises:

    a data support subunit configured to acquire a number of company names and perform a word segmentation, obtain a sequence of partial words, count the word frequency of each of the plurality of word segment sequences, and the adjacent phrase frequency of each pair of adjacent words, and calculate The closeness of each pair of adjacent words.
15. Apparatus according to any one of claims 11-14, wherein said data support subunit is further configured to mark high frequency words in each of said plurality of word segmentation sequences as institutional words.
16. The apparatus according to claim 15, wherein the high frequency word is filtered by: a word frequency of the high frequency word is not less than a third threshold, and a sum of word frequencies of each high frequency word and a word frequency of each end word The ratio of the sum is not less than the fourth threshold.
17. The device according to any one of claims 10-16, wherein the point of interest mining unit comprises:

    a filtering subunit configured to acquire a plurality of search information, and filter search information not including the brand word;

    Focusing on the word mining sub-unit, configured to cut the filtered search information, remove the brand words and then sort the words to select a number of high-frequency attention words;

    The point of interest labeling subunit is configured to label the points of interest of each of the attention words.
18. An advertisement placement device, comprising the advertisement placement keyword recommendation device according to any one of claims 10-17, and

    The delivery unit is configured to perform advertisement placement by using a combination of the recommended brand words and any attention words corresponding to the points of interest as keywords.
19. A device, characterized in that the device comprises:

    One or more processors;

    Memory for storing one or more programs,

    The one or more processors are caused to perform the method of any one of claims 1-9 when the one or more programs are executed by the one or more processors.
20. A computer readable storage medium storing a computer program, wherein the program is executed by a processor to implement the method of any one of claims 1-9.

Images