A search keyword mining method and device

Abstract

The application discloses a mining method and device for search keywords. The method comprises the following steps: obtaining a search behavior log of a user; identifying one or more groups of associated search keywords according to the search behavior log; and mining a search keyword to be optimized according to search results of the associated search keywords. The beneficial effect is that the result page of the user's dissatisfaction with the keyword search result can be quickly and automatically mined from the user behavior log, and the satisfaction degree of all users searching the search keyword is compared horizontally, so that the search keyword to be optimized is finally mined. The application provides strong support for improving the evaluation reputation of the search engine, the user use stickiness and active time, the overall user satisfaction and the like. Moreover, in the face of massive search results, the application adopts a machine automatic rapid processing mode, so that the problems of high cost and low efficiency under the manual processing mode are avoided, and the efficiency of the mining work is greatly improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of information mining, and in particular to a search keyword mining method and device. BACKGROUND

[0002] The user satisfaction index of the search engine search result will directly affect the evaluation of the search engine, and then affect the user's use stickiness and active time, so the user satisfaction index has important reference significance for search engine optimization. However, the user quantity and search result data quantity of the search engine are extremely large, and manual mining and optimization cannot meet the needs of practical application, so a method for quickly and automatically mining search keywords by using a machine is needed, which is applied to the field of search keyword mining to improve work efficiency and provide support for improving user experience. SUMMARY

[0003] In view of the above problems, the present application is proposed to provide a search keyword mining method and device which overcomes the above problems or at least partially solves the above problems.

[0004] According to one aspect of the present application, a search keyword mining method is provided, comprising:

[0005] obtaining a search behavior log of a user;

[0006] identifying one or more groups of associated search keywords according to the search behavior log;

[0007] mining a search keyword to be optimized according to the search results of the associated search keywords.

[0008] Optionally, the identifying one or more groups of associated search keywords according to the search behavior log comprises:

[0009] sorting the search behavior of the user according to the occurrence time according to the search behavior log;

[0010] sliding in a preset time window, and determining whether the search keywords corresponding to the two adjacent search behaviors in the time window are associated search keywords.

[0011] Optionally, the associated search keywords have a semantic association relationship, or the search keyword corresponding to the later search behavior is the name of the search engine.

[0012] Optionally, the mining a search keyword to be optimized according to the search results of the associated search keywords comprises:

[0013] determining whether the search result of the search keyword corresponding to the earlier search behavior meets at least one preset condition.

[0014] If yes, the search keyword corresponding to the search behavior occurring first is taken as the search keyword to be optimized.

[0015] Optionally, the preset condition comprises at least one of the following:

[0016] The number of search results of the first type in the search result homepage is greater than a preset threshold value;

[0017] The search result of the second type does not appear;

[0018] The search result of the second type is ranked outside a preset range;

[0019] The search result is not clicked.

[0020] Optionally, the search result of the first type is an advertisement.

[0021] Optionally, the search result of the second type is an application box onebox.

[0022] Optionally, the method further comprises:

[0023] Recording the number of times each search keyword to be optimized is mined;

[0024] Determining whether the ratio of the number of times to the total number of users searching for the search keyword is less than a preset threshold value, and if yes, marking the priority of the search keyword to be optimized as high, and if not, marking the priority as low.

[0025] According to another aspect of the present application, a mining device for search keywords is provided, comprising:

[0026] An acquisition unit adapted to acquire a search behavior log of a user;

[0027] An identification unit adapted to identify one or more groups of associated search keywords according to the search behavior log;

[0028] A mining unit adapted to mine search keywords to be optimized according to the search results of the associated search keywords.

[0029] Optionally, the identification unit is further adapted to sort the search behaviors of the user according to the occurrence time according to the search behavior log;

[0030] Sliding in a preset time window, and determining whether the search keywords corresponding to two adjacent search behaviors in the time window are associated search keywords.

[0031] Optionally, the associated search keywords have a semantic association relationship; or the search keyword corresponding to the search behavior occurring later is the name of a search engine.

[0032] Optionally, the mining unit is adapted to judge whether the search result of the search keyword corresponding to the search behavior occurring first meets at least one preset condition.

[0033] If yes, the search keyword corresponding to the search behavior occurring first is taken as a search keyword to be optimized.

[0034] Optionally, the preset condition comprises at least one of the following:

[0035] The number of search results of the first type in the search result homepage is greater than a preset threshold value;

[0036] The search result of the second type does not appear;

[0037] The search result of the second type is ranked outside a preset range;

[0038] The search result is not clicked.

[0039] Optionally, the search result of the first type is an advertisement.

[0040] Optionally, the search result of the second type is an application box onebox.

[0041] Optionally, the apparatus further comprises a marking unit adapted to record the number of times each search keyword to be optimized is mined;

[0042] Judge whether the ratio of the number of times to the total number of users searching for the search keyword is less than a preset threshold value, and if yes, mark the priority of the search keyword to be optimized as high, and if not, mark the priority as low.

[0043] According to still another aspect of the present application, there is provided an electronic device comprising: a processor; and a memory arranged to store computer executable instructions which, when executed, cause the processor to perform the method of any of the above.

[0044] According to yet another aspect of the present application, there is provided a computer readable storage medium, wherein the computer readable storage medium stores one or more programs which, when executed by a processor, implement the method of any of the above.

[0045] From the above, the technical scheme of the present application obtains the search behavior log of a user, identifies one or more groups of associated search keywords according to the search behavior log, and mines the search keyword to be optimized according to the search results of the associated search keywords. The beneficial effect is that the result page of which the user is not satisfied with the search result of the keyword can be quickly and automatically mined from the user behavior log, and the satisfaction degree of all users who search the keyword is compared horizontally, and finally the search keyword to be optimized is mined. This provides strong support for improving the evaluation reputation of the search engine, the user use stickiness and active time, the overall user satisfaction, and the like. Moreover, in the face of massive search results, the present application adopts a machine automatic and rapid processing mode, avoids the problems of high cost and low efficiency in the manual processing mode, and greatly improves the efficiency of the mining work.

[0046] The above description is only a summary of the technical scheme of the present application, in order to more clearly understand the technical means of the present application, and the present application can be implemented according to the content of the specification, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific embodiments of the present application are described below. BRIEF DESCRIPTION OF DRAWINGS

[0047] 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 embodiments. The accompanying drawings are included to provide a description of the preferred embodiments and are not meant to limit the present application. Furthermore, the same reference numerals are intended to denote the same components throughout the accompanying drawings. In the drawings:

[0048] Figure 1 A flowchart of a search result mining method according to one embodiment of the present application is shown;

[0049] Figure 2 A structural diagram of a search result mining device according to one embodiment of the present application is shown;

[0050] Figure 3 A structural diagram of an electronic device according to one embodiment of the present application is shown;

[0051] Figure 4 A structural diagram of a computer readable storage medium according to one embodiment of the present application is shown. DETAILED DESCRIPTION

[0052] Exemplary embodiments of the present disclosure will be described in greater detail below with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it is understood that the present disclosure can be embodied in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the present disclosure to those skilled in the art.

[0053] Figure 1 A flowchart of a search result mining method according to an embodiment of the present disclosure is shown. As shown, the method includes: Figure 1

[0054] In step S110, search behavior logs of a user are obtained.

[0055] When a user searches for a keyword using a search engine, the search engine records the user's search behavior. The search behavior logs can include, for example, records of the search keyword, information of the search result page, access time, etc. By obtaining these search behavior logs, the specific information of the user's search behavior in the search engine can be intuitively reflected.

[0056] In step S120, one or more groups of associated search keywords are identified according to the search behavior logs.

[0057] When a user obtains relevant information using a search engine, different search behaviors can be exhibited due to different satisfaction levels of the result pages for different search keywords. After searching for a specific keyword in the search engine, if a satisfactory search result page is obtained, there can be no subsequent associated search behavior. If an accurate search result is not obtained, the search strategy can be adjusted, and then another search keyword can be searched, for example, a user searches for "Peking University" and does not obtain useful information, and then adjusts the search keyword to "Peking University official website" and searches again until the user obtains information that is considered to be accurate. Therefore, by analyzing several groups of search keywords that have relevance in the search behavior, the satisfaction level of the user for the search keyword result page can be reflected.

[0058] In step S130, a search keyword to be optimized is mined according to the search results of the associated search keywords.

[0059] According to the user's behavior representation for a specific keyword, a search keyword to be optimized is identified and marked, and is assisted by artificial comprehensive analysis and confirmation, etc., so as to further optimize the search result, thereby improving the user experience and enhancing the user satisfaction index.

[0060] As can be seen, Figure 1 ​The method can quickly and automatically mine the result pages that users are not satisfied with from the user search behavior log, and through horizontal comparison of the satisfaction of all users who search the search keyword, finally mine the search keyword to be optimized. The method provides strong support for improving the evaluation reputation of the search engine, user use stickiness and active time, user overall satisfaction, and the like. Moreover, in the face of massive search results, the method adopts a machine automatic and rapid processing mode, avoids the problems of high cost and low efficiency in the manual processing mode, and thus greatly improves the efficiency of the mining work.

[0061] In an embodiment of the present application, in the above method, the identifying one or more groups of associated search keywords from the search behavior log comprises: sorting the search behaviors of the user according to the occurrence time according to the search behavior log; sliding in a preset time window, and determining whether the search keywords corresponding to two adjacent search behaviors in the time window are associated search keywords.

[0062] The information list in the search behavior log can include the record of the search keyword, the search result page URL information, the access time, and the like. The search behavior log list is reordered according to the occurrence time of the search behavior, and a new search behavior log table arranged according to the search behavior occurrence time relationship is obtained. Then, a time window is preset, the time interval can be set to 10 minutes, and it is determined whether there is an association relationship between the search keywords corresponding to two adjacent search behaviors in the 10 minutes. The preset 10-minute time window is smoothly slid in the new search behavior log table, and it is determined whether there is an association relationship between the search keywords in the time window during the sliding process, until the search behavior log is completely covered. After the sorting and the determination in the preset time window, the search behavior log list can more clearly reflect the search behavior characteristics of the user in a certain time segment, and lays a foundation for the subsequent accurate mining work.

[0063] In an embodiment of the present application, in the above method, the semantic of the associated search keywords has an association relationship; or the search keyword corresponding to the later-occurring search behavior is the name of the search engine.

[0064] If the user uses the keywords with the association relationship as the search keywords in front and back respectively, the behavior characteristics inherently contain the satisfaction degree of the user for the obtained search results, and therefore it is necessary to explicitly have the search keywords with the association relationship in the search behavior log. The judgment of whether there is the association relationship can be analyzed from the semantic association degree. For example, the search keywords "Beijing University" and "Beijing University official website" and "Beida official website" have the semantic association. There is another case that the user searches a keyword first and then searches the name of other search engines outside the search engine as the keyword, and at this time, the search keywords in front and back are considered to have the association relationship.

[0065] In an embodiment of the present application, in the above method, the mining of the search keywords to be optimized according to the search results of the associated search keywords comprises: judging whether the search results of the search keywords corresponding to the search behavior occurring first meet at least one preset condition; if yes, taking the search keywords corresponding to the search behavior occurring first as the search keywords to be optimized.

[0066] For example, a condition is preset, when the search keyword searched first is "Beijing University", it is analyzed and judged whether the search results corresponding to the search keyword "Beijing University" meet the preset condition. Once it is met, it is determined that the search keyword "Beijing University" is the search keyword which needs to be further optimized, thereby realizing the mining of the search keywords to be optimized.

[0067] In an embodiment of the present application, in the above method, the preset condition comprises at least one of the following: the number of the search results of the first type in the first page of the search results is greater than a preset threshold; the search results of the second type do not appear; the search results of the second type are ranked outside a preset range; the search results are not clicked.

[0068] The display effect of the search result page directly affects the user satisfaction, if the number of a certain type of search results is too much and the association degree is small, it will directly affect the user experience. Therefore, a certain threshold is preset, which can be set to 5, and once the threshold is exceeded, the corresponding search keyword will be determined as the search keyword to be optimized. Certain types of search results need to be displayed within a certain range to achieve better user experience, and for this type of search result, a certain display range needs to be preset, which can be the sequence position within the first 3 positions, and if it exceeds this range, the corresponding search keyword will be determined as the search keyword to be optimized. If the search results appear, but the user does not click the link of the search results, in this case, it is also presumed that the user is not satisfied with the search results, and the corresponding search keyword will be determined as the search keyword to be optimized. In this way, the user satisfaction index is described by a quantitative index, and the accuracy of the mining is improved.

[0069] In one embodiment of the present invention, in the above method, the first type of search result is an advertisement.

[0070] An appropriate number of ads has little impact on user experience, but when the number of ads exceeds a certain level, it will have a significant impact on user satisfaction, and search result satisfaction will drop sharply.

[0071] In one embodiment of the present invention, in the above method, the second type of search result is the application box Onebox.

[0072] Onebox search results are a unique type of display that can include various vertical information such as news, weather, time, travel, and ticketing. The background of this type of search result display often has distinctive features, making it easy to identify. If it's displayed too far back, users may not see it, negating the purpose of highlighting it and significantly impacting user satisfaction. Therefore, onebox search results should be displayed as early as possible in the search results.

[0073] In one embodiment of the present invention, the method further includes: recording the number of times each search keyword to be optimized is mined; determining whether the ratio of the number of times to the total number of users who have searched for the search keyword is less than a preset threshold; if so, marking the priority of the search keyword to be optimized as high, otherwise marking it as low.

[0074] To determine whether satisfaction levels for specific search keywords reflect individual user preferences or general trends, it's necessary to analyze individual user satisfaction levels followed by a comparative analysis of overall satisfaction levels. This helps identify search keywords with generally low satisfaction rates. For example, for the search keyword "Peking University," the number of times each user's search for this keyword is recorded, along with the total number of users who have searched for it. The ratio of these recorded searches to the total number of users searching for this keyword is then compared to a preset threshold. This threshold could be 0.9. If the ratio is less than 0.9, the keyword "Peking University" is marked as having a high optimization priority. If the ratio is greater than or equal to 0.9, it is marked as having a low optimization priority. Subsequent optimization efforts can then proceed in an orderly manner based on this priority level. This approach effectively prevents potential biases in the search results and improves the accuracy of the analysis.

[0075] Figure 2 A schematic diagram of a search result mining device according to an embodiment of the present invention is shown, as follows: Figure 2 As shown, the device 200 includes:

[0076] Acquisition unit 210 is adapted to acquire user search behavior logs;

[0077] When users search for keywords using a search engine, the search engine records their behavior. This search behavior log can include information such as search keywords, search result page URLs, and access times. By obtaining these search behavior logs, specific information about a user's search behavior within the search engine can be clearly observed.

[0078] The identification unit 220 is adapted to identify one or more sets of related search keywords based on search behavior logs;

[0079] When users obtain relevant information through search engines, their subsequent behavior may vary depending on their satisfaction with the search results pages for different search keywords. If a user finds a satisfactory search result page after searching for a specific keyword, they may not engage in further related searches. However, if they do not find accurate results, they may adjust their search strategy and search for another keyword. For example, a user searching for information on Peking University's official website might initially search for "Peking University" but find nothing useful. They might then change their search to "Peking University official website" and search again until they find information they deem relatively accurate. Therefore, analyzing several related groups of search keywords in these search behaviors can reflect users' satisfaction with the search result pages.

[0080] Mining unit 230 is adapted to mine search keywords to be optimized based on search results of related search keywords.

[0081] Based on users' behavioral patterns toward specific keywords, we identify and mark search keywords that need optimization. With the assistance of manual comprehensive analysis and confirmation, we can further optimize search results, thereby improving user experience and increasing user satisfaction metrics.

[0082] It can be seen that, as Figure 2 The device shown can quickly and automatically extract result pages from user search behavior logs that users are dissatisfied with for certain keywords. By comparing the satisfaction levels of all users who have searched for the same keyword, it ultimately identifies search keywords that need optimization. This provides strong support for improving search engine reputation, user stickiness and activity time, and overall user satisfaction. Moreover, when faced with massive amounts of search results, this invention adopts a rapid, automated machine processing method, avoiding the high costs and low efficiency of manual processing, thereby significantly improving the efficiency of the data extraction work.

[0083] In one embodiment of the present invention, the identification unit is further adapted to sort the user's search behavior according to the occurrence time based on the search behavior log; and to slide the search behavior through a preset time window to determine whether the search keywords corresponding to two adjacent search behaviors in the time window are related search keywords.

[0084] The search behavior log may contain records of search keywords, search result page URLs, and access times. The search behavior log list is reordered based on the time of the search behavior, resulting in a new search behavior log table arranged chronologically. Then, a preset time window, such as a 10-minute interval, is used to determine if there is a correlation between the search keywords corresponding to two adjacent search behaviors within this 10-minute period. This preset 10-minute time window is then smoothly slid across the new search behavior log table, checking for correlations between search keywords within the time window each time, until the entire search behavior log is covered. The search behavior log list, after being sorted and processed within the preset time window, will more clearly reflect the characteristics of user search behavior within a certain time segment, laying the foundation for subsequent accurate data mining.

[0085] In one embodiment of the present invention, in the above-described apparatus, the semantics of the associated search keywords are related; or, the search keyword corresponding to the subsequent search behavior is the name of the search engine.

[0086] If a user uses related keywords in their searches, their behavior inherently reflects their satisfaction with the search results. Therefore, it's necessary to clearly identify related keywords in the search behavior log. Determining the existence of a relationship can be done by analyzing semantic relevance. For example, the search keyword "Peking University" is semantically related to "Peking University official website" and "PKU official website." Another scenario is when a user searches for one keyword and then searches using the name of another search engine; in this case, the two searches are also considered related.

[0087] In one embodiment of the present invention, in the above-described apparatus, the mining unit is adapted to determine whether the search results of the search keywords corresponding to the preceding search behavior meet at least one preset condition; if so, the search keywords corresponding to the preceding search behavior are taken as the search keywords to be optimized.

[0088] For example, a preset condition can be set: when the first search keyword is "Peking University", the analysis will determine whether the search results for "Peking University" meet the preset condition. If they do, "Peking University" is identified as a search keyword that needs further optimization, thus achieving the work of discovering search keywords to be optimized.

[0089] In one embodiment of the present invention, the preset conditions in the above-described device include at least one of the following: the number of search results of the first type on the search results homepage is greater than a preset threshold; no search results of the second type appear; the search results of the second type are ranked outside a preset range; and the search results are not clicked.

[0090] The display effect of search results pages directly impacts user satisfaction. An excessive number of search results of a certain type with low relevance will negatively affect user experience. Therefore, a threshold of 5 results is preset. Once this threshold is exceeded, the corresponding search keywords are identified as keywords to be optimized. Certain types of search results require a specific display range for optimal user experience. For these types of results, a preset display range is needed, which could be the sequence position within the first 3 results. Search keywords exceeding this range are identified as keywords to be optimized. If a search result appears but the user does not click on the link, it is presumed that the user is dissatisfied with the search result, and the corresponding search keyword is identified as a keyword to be optimized. In this way, user satisfaction metrics are described through quantitative indicators, improving the accuracy of the analysis.

[0091] In one embodiment of the present invention, in the above-described apparatus, the first type of search result is an advertisement.

[0092] An appropriate number of ads has little impact on user experience, but when the number of ads exceeds a certain level, it will have a significant impact on user satisfaction, and search result satisfaction will drop sharply.

[0093] In one embodiment of the present invention, in the above-described device, the second type of search result is the application box Onebox.

[0094] Onebox search results are a unique type of display that can include various vertical information such as news, weather, time, travel, and ticketing. The background of this type of search result display often has distinctive features, making it easy to identify. If it's displayed too far back, users may not see it, negating the purpose of highlighting it and significantly impacting user satisfaction. Therefore, onebox search results should be displayed as early as possible in the search results.

[0095] In one embodiment of the present invention, the device further includes a marking unit 240, which is adapted to record the number of times each search keyword to be optimized is mined; and to determine whether the ratio of the number of times to the total number of users who have searched for the search keyword is less than a preset threshold. If so, the priority of the search keyword to be optimized is marked as high; otherwise, it is marked as low.

[0096] To determine whether satisfaction levels for specific search keywords reflect individual user preferences or general trends, it's necessary to analyze individual user satisfaction levels followed by a comparative analysis of overall satisfaction levels. This helps identify search keywords with generally low satisfaction rates. For example, for the search keyword "Peking University," the number of times each user's search for this keyword is recorded, along with the total number of users who have searched for it. The ratio of these recorded searches to the total number of users searching for this keyword is then compared to a preset threshold. This threshold could be 0.9. If the ratio is less than 0.9, the keyword "Peking University" is marked as having a high optimization priority. If the ratio is greater than or equal to 0.9, it is marked as having a low optimization priority. Subsequent optimization efforts can then proceed in an orderly manner based on this priority level. This approach effectively prevents potential biases in the search results and improves the accuracy of the analysis.

[0097] It should be noted that the specific implementation methods of the above-mentioned device embodiments can be referred to the specific implementation methods of the corresponding methods described above, and will not be repeated here.

[0098] In summary, the technical solution of this invention involves acquiring user search behavior logs; identifying one or more sets of related search keywords based on the search behavior logs; and mining search keywords to be optimized based on the search results of the related search keywords. The beneficial effect of this technical solution is that it can quickly and automatically mine result pages from user search behavior logs where users are dissatisfied with the keyword search results, and by horizontally comparing the satisfaction levels of all users who have searched for that keyword, it ultimately mines search keywords to be optimized. This provides strong support for improving search engine reputation, user stickiness and activity time, and overall user satisfaction. Moreover, facing massive amounts of search results, this invention adopts a rapid, automated machine processing method, avoiding the high costs and low efficiency of manual processing, thereby significantly improving the efficiency of the mining work.

[0099] It should be noted that:

[0100] The algorithms and displays provided herein are not inherently related to any particular computer, virtual device, or other equipment. Various general-purpose devices can also be used in conjunction with the teachings herein. The required structure for constructing such devices is apparent from the above description. Furthermore, this invention is not directed to any particular programming language. It should be understood that the contents of the invention described herein can be implemented using various programming languages, and the above description of specific languages ​​is for the purpose of disclosing the best mode of implementation of the invention.

[0101] Numerous specific details are set forth in the specification provided herein. However, it will be 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 so as not to obscure the understanding of this specification.

[0102] Similarly, it should be understood that, in order to simplify this disclosure and aid in understanding one or more of the various aspects of the invention, in the above description of exemplary embodiments of the invention, various features of the invention are sometimes grouped together in a single embodiment, figure, or description thereof. However, this method of disclosure should not be construed as reflecting an intention that the claimed invention requires more features than are expressly recited in each claim. Rather, as reflected in the following claims, inventive aspects lie in fewer than all features of a single foregoing disclosed embodiment. Therefore, the claims following the detailed description are hereby expressly incorporated into this detailed description, wherein each claim itself is a separate embodiment of the invention.

[0103] Those skilled in the art will understand that modules in the device of the embodiments can be adaptively changed and placed in one or more devices different from that embodiment. Modules, units, or components in the embodiments can be combined into a single module, unit, or component, and further, they can be divided into multiple sub-modules, sub-units, or sub-components. Except where at least some of such features and / or processes or units are mutually exclusive, any combination can be used to combine all features disclosed in this specification (including the accompanying claims, abstract, and drawings) and all processes or units of any method or device so disclosed. Unless expressly stated otherwise, each feature disclosed in this specification (including the accompanying claims, abstract, and drawings) may be replaced by an alternative feature that serves the same, equivalent, or similar purpose.

[0104] Furthermore, those skilled in the art will understand that although some embodiments described herein include certain features but not others included in other embodiments, combinations of features from different embodiments are intended to be within the scope of the invention and form different embodiments. For example, in the following claims, any of the claimed embodiments can be used in any combination.

[0105] The various component embodiments of the present invention can be implemented in hardware, or as software modules running on one or more processors, or a combination thereof. Those skilled in the art will understand that microprocessors or digital signal processors (DSPs) can be used in practice to implement some or all of the functions of some or all of the components in the keyword mining apparatus according to embodiments of the present invention. The present invention can also be implemented as a device or apparatus program (e.g., a computer program and computer program product) for performing part or all of the methods described herein. Such programs implementing the present invention can be stored on a computer-readable medium or can be in the form of one or more signals. Such signals can be downloaded from an Internet website, provided on a carrier signal, or provided in any other form.

[0106] For example, Figure 3 A schematic diagram of an electronic device according to an embodiment of the present invention is shown. The electronic device includes a processor 310 and a memory 320 arranged to store computer-executable instructions (computer-readable program code). The memory 320 may be an electronic memory such as flash memory, EEPROM (Electrically Erasable Programmable Read-Only Memory), EPROM, hard disk, or ROM. The memory 320 has a storage space 330 for storing computer-readable program code 331 for performing any of the method steps described above. For example, the storage space 330 for storing computer-readable program code may include various computer-readable program codes 331 respectively for implementing the various steps in the methods described above. The computer-readable program code 331 can be read from or written to one or more computer program products. These computer program products include program code carriers such as hard disks, CDs, memory cards, or floppy disks. Such computer program products are typically, for example... Figure 4 The aforementioned computer-readable storage medium. Figure 4A schematic diagram of a computer-readable storage medium according to an embodiment of the present invention is shown. The computer-readable storage medium 400 stores computer-readable program code 331 for performing the method steps according to the present invention, which can be read by the processor 310 of an electronic device 300. When the computer-readable program code 331 is executed by the electronic device 300, it causes the electronic device 300 to perform the various steps of the method described above. Specifically, the computer-readable program code 331 stored in the computer-readable storage medium can perform the methods shown in any of the above embodiments. The computer-readable program code 331 can be compressed in a suitable form.

[0107] It should be noted that the above embodiments are illustrative of the invention and not restrictive, and that those skilled in the art can devise alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses should not be construed as limiting the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. 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 different elements and by means of a suitably programmed computer. In the unit claims enumerating several means, several of these means may be embodied by 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.

[0108] Embodiments of the present invention disclose A1, a method for mining search keywords, comprising:

[0109] Obtain user search behavior logs;

[0110] Based on the search behavior logs, one or more sets of related search keywords are identified;

[0111] Based on the search results of the associated search keywords, search keywords to be optimized are extracted.

[0112] A2. The method as described in A1, wherein identifying one or more sets of related search keywords based on the search behavior log includes:

[0113] The user's search behavior is sorted by the time it occurred based on the search behavior log;

[0114] The algorithm slides within a preset time window to determine whether the search keywords corresponding to two adjacent search behaviors within the time window are related search keywords.

[0115] A3. As described in A2, wherein the semantics of the associated search keywords are related; or, the search keyword corresponding to the subsequent search behavior is the name of the search engine.

[0116] A4. The method as described in A2, wherein mining the search keywords to be optimized based on the search results of the associated search keywords includes:

[0117] Determine whether the search results for the search keywords corresponding to the preceding search behavior meet at least one preset condition;

[0118] If so, the search keywords corresponding to the first search behavior will be used as the search keywords to be optimized.

[0119] A5. The method as described in A4, wherein the preset conditions include at least one of the following:

[0120] On the first page of search results, the number of search results of the first type exceeds a preset threshold;

[0121] No search results of the second type appeared;

[0122] The second type of search results are ranked outside the preset range;

[0123] The search results were not clicked.

[0124] A6. The method described in A5, wherein the search results of the first type are advertisements.

[0125] A7. As described in A5, wherein the search results of the second type are application boxes (onebox).

[0126] A8. The method as described in A1, wherein the method further includes:

[0127] Record the number of times each search keyword to be optimized was discovered;

[0128] If the ratio of the number of searches to the total number of users who have searched for the keyword is less than a preset threshold, the keyword to be optimized is marked as high priority; otherwise, it is marked as low priority.

[0129] Embodiments of the present invention also disclose B9, a keyword mining device, comprising:

[0130] The acquisition unit is suitable for acquiring users' search behavior logs;

[0131] The identification unit is adapted to identify one or more sets of related search keywords based on the search behavior log;

[0132] The mining unit is adapted to mine search keywords to be optimized based on the search results of the associated search keywords.

[0133] B10. The apparatus as described in B9, wherein...

[0134] The identification unit is adapted to sort the user's search behavior by occurrence time according to the search behavior log; and slide it in a preset time window to determine whether the search keywords corresponding to two adjacent search behaviors in the time window are related search keywords.

[0135] B11. The apparatus as described in B10, wherein the semantics of the associated search keywords are related; or, the search keyword corresponding to the subsequent search behavior is the name of the search engine.

[0136] B12. The apparatus as described in B10, wherein...

[0137] The mining unit is adapted to determine whether the search results of the search keywords corresponding to the preceding search behavior meet at least one preset condition; if so, the search keywords corresponding to the preceding search behavior are taken as the search keywords to be optimized.

[0138] B13. The apparatus as described in B12, wherein the preset conditions include at least one of the following:

[0139] On the first page of search results, the number of search results of the first type exceeds a preset threshold;

[0140] No search results of the second type appeared;

[0141] The second type of search results are ranked outside the preset range;

[0142] The search results were not clicked.

[0143] B14. The apparatus as described in B13, wherein the search results of the first type are advertisements.

[0144] B15. The device as described in B13, wherein the search results of the second type are application boxes (onebox).

[0145] B16. The apparatus as described in B9, wherein the apparatus further comprises:

[0146] The marking unit is adapted to record the number of times each search keyword to be optimized is mined; it determines whether the ratio of the number of times to the total number of users who have searched for the search keyword is less than a preset threshold. If so, the priority of the search keyword to be optimized is marked as high; otherwise, it is marked as low.

[0147] Embodiments of the present invention also disclose C17, an electronic device comprising: a processor; and a memory arranged to store computer-executable instructions, which, when executed, cause the processor to perform a method as described in any one of A1-A8.

[0148] Embodiments of the present invention also disclose D18, a computer-readable storage medium, wherein the computer-readable storage medium stores one or more programs, which, when executed by a processor, implement the method as described in any one of A1-A8.

Claims

1. A method for mining search keywords, comprising: Obtain user search behavior logs; Based on the search behavior logs, one or more sets of related search keywords are identified; Based on the search results of the associated search keywords, identify the search keywords to be optimized; The process of mining the search keywords to be optimized based on the search results of the associated search keywords includes: Determine whether the search results for the search keywords corresponding to the preceding search behavior meet at least one preset condition; If so, the search keywords corresponding to the first search behavior will be used as the search keywords to be optimized; The preset conditions include at least one of the following: the number of search results of the first type on the search results homepage is greater than a preset threshold; no search results of the second type appear; the search results of the second type are not sorted within a preset range; and the search results are not clicked.

2. The method as described in claim 1, wherein, The identification of one or more sets of related search keywords based on the search behavior log includes: The user's search behavior is sorted by the time it occurred based on the search behavior log; The algorithm slides within a preset time window to determine whether the search keywords corresponding to two adjacent search behaviors within the time window are related search keywords.

3. The method as described in claim 2, wherein, The related search keywords have a semantic relationship; or, the search keyword corresponding to the subsequent search behavior is the name of the search engine.

4. The method of claim 1, wherein, The first type of search result is an advertisement.

5. The method of claim 1, wherein, The second type of search result is the application box Onebox.

6. The method of claim 1, wherein, The method further includes: Record the number of times each search keyword to be optimized was discovered; If the ratio of the number of searches to the total number of users who have searched for the keyword is less than a preset threshold, the keyword to be optimized is marked as high priority; otherwise, it is marked as low priority.

7. A device for mining search keywords, comprising: The acquisition unit is suitable for acquiring users' search behavior logs; The identification unit is adapted to identify one or more sets of related search keywords based on the search behavior log; The mining unit is adapted to mine search keywords to be optimized based on the search results of the associated search keywords; The mining unit is adapted to determine whether the search results of the search keywords corresponding to the preceding search behavior meet at least one preset condition; if so, the search keywords corresponding to the preceding search behavior are taken as the search keywords to be optimized. The preset conditions include at least one of the following: the number of search results of the first type on the search results homepage is greater than a preset threshold; no search results of the second type appear; the search results of the second type are not sorted within a preset range; and the search results are not clicked.

8. The apparatus of claim 7, wherein, The identification unit is adapted to sort the user's search behavior according to the time of occurrence based on the search behavior log; The algorithm slides within a preset time window to determine whether the search keywords corresponding to two adjacent search behaviors within the time window are related search keywords.

9. The apparatus of claim 8, wherein, The related search keywords have a semantic relationship; or, the search keyword corresponding to the subsequent search behavior is the name of the search engine.

10. The apparatus of claim 7, wherein, The first type of search result is an advertisement.

11. The apparatus of claim 7, wherein, The second type of search result is the application box Onebox.

12. The apparatus of claim 7, wherein, The device further includes: The marking unit is adapted to record the number of times each search keyword to be optimized is mined; it determines whether the ratio of the number of times to the total number of users who have searched for the search keyword is less than a preset threshold. If so, the priority of the search keyword to be optimized is marked as high; otherwise, it is marked as low.

13. An electronic device, wherein, The electronic device includes: a processor; and a memory arranged to store computer-executable instructions, which, when executed, cause the processor to perform the method as described in any one of claims 1-6.

14. A computer-readable storage medium, wherein, The computer-readable storage medium stores one or more programs that, when executed by a processor, implement the method as described in any one of claims 1-6.

Images