Website navigation service system based on artificial intelligence

Abstract

The invention discloses a website navigation method based on artificial intelligence, and the method comprises the steps: extracting entry information currently searched by a front-end page user, historical search record data of the user and ip address data of the user, carrying out the keyword division of the obtained historical search record data, obtaining historical keywords, and storing the historical keywords in a database; the method comprises the following steps: searching entry information, processing according to the searched entry information to obtain entry keywords, retrieving associated keywords in a system library according to the entry keywords, semantically integrating each keyword and the associated keywords into a long-tail keyword, searching according to the long-tail keyword, and evaluating the entry information through an evaluation algorithm. According to the method, the search items are sorted according to the analysis result, the final result is sent to the search page, the search result is automatically fed back to the back-end server, the user feeds back the opinion of the search result based on the opinion interface established by the background server, and the retrieval accuracy and the user satisfaction degree are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of website navigation services, and in particular to an artificial intelligence-based website navigation service system. Background Art

[0002] Nowadays, artificial intelligence website navigation system services play an important role in helping users solve problems. By providing rich AI tools and resources, these navigation systems can meet the needs of different users, from AI chat assistants to image generation, writing assistance, and business management. However, despite the convenience brought by these services, the existing artificial intelligence website navigation system encounters broader and more ambiguous words based on the concept of keyword matching technology, resulting in inaccurate search results. That is, when the user fails to use the most suitable keywords to describe the query requirements, or enters a very broad vocabulary, it will make it difficult for the search engine to match the most relevant results. Based on this phenomenon, the search results are inaccurate, which ultimately affects the user experience and reduces the user's satisfaction and dependence on the website navigation system. Therefore, it is necessary to design an artificial intelligence-based website navigation service system that can better improve the accuracy of retrieval and user satisfaction. Summary of the invention

[0003] The purpose of the present invention is to provide a website navigation service system based on artificial intelligence to solve the problems raised in the above background technology.

[0004] In order to solve the above technical problems, the present invention provides the following technical solutions: an artificial intelligence-based website navigation service system, which executes an artificial intelligence-based website navigation method, and the operating steps of the method include:

[0005] According to the above technical solution, step S1: extracting the term information currently searched by the user on the front-end page, as well as the user's historical search record data and the user's IP address data;

[0006] Step S2: dividing the obtained historical search record data into keywords to obtain historical keywords, storing them in a backend database, processing them according to the searched term information to obtain term keywords, and then searching for related keywords in the system library according to the term keywords;

[0007] Step S3: semantically integrate each keyword with the associated keywords to obtain a long-tail keyword array, and search based on these long-tail keywords;

[0008] Step S4: sorting the search terms according to the results of the system analysis through the search engine evaluation algorithm, and sending the final results to the search page;

[0009] Step S5: The search results are automatically fed back to the backend server, and the user provides feedback on the search results based on the opinion interface built by the backend server.

[0010] According to the above technical solution, step S2 further includes the following steps:

[0011] Step S21, the search engine performs keyword processing based on the user's historical search record data to obtain the historical keywords. When weighting the historical keywords, the more recent the appearance time, the greater the weight of the historical keyword, and the more frequent the appearance, the greater the weight of the historical keyword. The weight value of the historical keyword division is calculated by the formula:

[0012]

[0013] S f =n i

[0014] S=αS t +(1-α)S f

[0015] Where m represents the time decay factor and decreases with time, t represents the current time, and t i Indicates the time when the keyword last appeared, T represents a time constant, S t represents the time weight, n i Indicates the number of times the keyword appears, S f represents the frequency weight, α represents a parameter for adjusting the importance of time weight and frequency weight, and S represents the final weight of the historical keyword;

[0016] Step S22: After obtaining the historical keyword, the backend database will attach multiple tag indexes to the historical keyword according to the associated scope words of the historical keyword to construct a multi-dimensional historical keyword array;

[0017] Step S23, the search engine divides the searched term information into the term keywords, and uses the PageRank algorithm to evaluate the weight of each term keyword, and divides them into the first term keyword and the second term keyword. The lower the weight of the keyword, the lower the order, and they are identified with numbers according to the order of the word sequence. However, when the position of a term keyword is swapped with another term keyword and the semantics remain unchanged, the above two term keywords are identified with the same number, and an array of term keywords with subscript identifiers will be obtained. Then, according to these keywords, related keywords are retrieved in the system library to obtain a multi-dimensional array of related keywords corresponding to each term keyword.

[0018] According to the above technical solution, step S23 further includes the following steps:

[0019] Step S231: When the search engine divides the entry keyword, if there are more than two entry keywords, the search engine will match the entry keyword with the historical keyword analyzed in step S21, and determine the matching value between the entry keyword and the historical keyword by semantic analysis technology. When the matching value reaches a set threshold, the historical keyword is used as the entry keyword of the searched entry information;

[0020] Step S232: On the contrary, when the search engine is dividing the entry keywords, if only one or two of the entry keywords are obtained, resulting in an unclear search range, step S221 will be repeated to determine whether there are relevant historical keywords that match them;

[0021] Step S233: The search engine attaches multiple tag indexes to the entry keyword according to the associated scope words of the entry keyword, constructs a multidimensional entry keyword array, and calculates the Mahalanobis distance d between the multidimensional entry keyword array and the multidimensional historical keyword array through the multidimensional historical keyword array analyzed in step S22:

[0022]

[0023] In the formula, α represents a multidimensional entry keyword array, β represents a multidimensional historical keyword array, ∑ -1 represents the inverse matrix of the covariance matrix;

[0024] The calculated Mahalanobis distance is compared with the preset value, and the historical keywords corresponding to the distance less than the preset condition are used as the associated keywords of the entry keyword, and the historical keywords that meet the conditions are sorted according to the closer the distance d, the greater the weight.

[0025] According to the above technical solution, step S3 further includes the following steps:

[0026] Step S31, the search engine calculates the Mahalanobis distance between the analyzed multidimensional entry keyword array and the multidimensional associated keyword array, and sorts the calculated Mahalanobis distance results in ascending order to obtain a distance sorting table, selects the top-ranked associated keywords, and makes the number of the entry keyword identification consistent with the number to be identified for the corresponding associated keyword;

[0027] Step S32, sequentially and orderly semantically integrate the corresponding entry keywords and the associated keywords in the order of sorting, perform semantic association between the entry keywords and the associated keywords according to the digital identifier in the integration basis, obtain the long-tail keyword array, and weight sort the long-tail keywords according to the Mahalanobis distance result of step S31;

[0028] Step S33, after obtaining the long-tail keyword array, compare and analyze it with the preset content of the back-end database to determine whether the user IP address data is needed to further divide the weight of the long-tail keyword. If it is determined to be yes, the search engine will dynamically adjust the weight of the long-tail keyword based on the user IP address data, or provide more relevant localized search results.

[0029] According to the above technical solution, step S4 further includes: the search engine ranks the long-tail keywords based on the weights obtained by analysis, the higher the ranking, the higher the matching degree with the term searched by the user, the closer the long-tail keyword with a high matching degree in the search item, the higher the ranking, and the smaller the corresponding item number.

[0030] According to the above technical solution, step S5 further includes: providing a feedback button or link on the search results page to allow users to evaluate or make suggestions on the search results, automatically triggering the optimization of the search analysis algorithm based on the feedback information, and using JavaScript to monitor the user's search behavior, monitor the user's search habits entered on the front-end page, the click-through rate of the search results, the specific items clicked, and the time spent on each search result, and use the above data as samples to train and improve the search engine's keyword extraction algorithm and adjust the weights of related keywords.

[0031] According to the above technical scheme, the website navigation service system based on artificial intelligence includes a data acquisition module, a keyword extraction module, a keyword aggregation module, a demand matching module and a result feedback module. The data acquisition module is used to obtain the term information currently searched by the user, as well as the user's historical search record data and the user's IP address data. The keyword extraction module is used to extract the term information currently searched by the user, the keywords of the user's historical search record data and related keywords. The keyword aggregation module is used to semantically integrate the acquired keywords to form long-tail keywords. The demand matching module is used to match the user's search needs according to the weight of its long-tail keywords and the user's IP address data.

[0032] According to the above technical scheme, the keyword extraction module includes a multidimensional historical keyword module, a multidimensional entry keyword module, and a multidimensional associated keyword module. The multidimensional historical keyword module is used to analyze the user's search habits and locate the user's search intentions. The multidimensional entry keyword module is used to extract keywords from the user's current search entry information and divide its keyword range to facilitate matching with the historical keyword module. The multidimensional associated keyword module is used to further expand the keywords of the entry keywords and add associated keywords.

[0033] According to the above technical scheme, the keyword extraction module includes a multidimensional historical keyword module, a multidimensional entry keyword module, and a multidimensional associated keyword module. The multidimensional historical keyword module is used to analyze the user's search habits and locate the user's search intentions. The multidimensional entry keyword module is used to extract keywords from the user's current search entry information and divide its keyword range to facilitate matching with the historical keyword module. The multidimensional associated keyword module is used to further expand the keywords of the entry keywords and add associated keywords.

[0034] According to the above technical solution, the keyword aggregation module includes a long-tail keyword module and a demand matching module. The long-tail keyword module is used to perform phonetic integration of the keywords analyzed by the keyword extraction module, and the demand matching module is used to sort the integrated long-tail keywords by calculated weight values, and match the user search needs in combination with the acquired user IP address data.

[0035] According to the above technical solution, the result feedback module includes a website information automatic update module and a background data maintenance module. The website information automatic update module is used to ensure the real-time nature of the website content and provide users with the latest search results. The background data maintenance module is used to monitor the user's search behavior and the feedback data actively provided by the user to adjust the background sample data.

[0036] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: the present invention, by acquiring the historical records and IP address data of user searches, analyzes the user's search habits and further accurately determines the user's search scope, extracts historical keywords from the historical records of user searches, and constructs a multidimensional historical keyword array by calculating the weight of each historical keyword and attaching a corresponding tag index. When facing the entry information searched by the user, the system will also extract its keywords and compare it with the back-end database to find the corresponding associated keywords, and will also construct a multidimensional associated keyword array. In addition, a multidimensional entry keyword array is constructed according to the associated scope words of the entry, and the weight value of the keyword is calculated by calculating the Mahalanobis distance between the multidimensional data groups to obtain the keyword and the associated keyword closest to the user's search intention, and the analysis results are semantically integrated to obtain long-tail keywords, and combined with the user's IP address data, the user's search intention can be better clarified, thereby improving the website's traffic quality and user satisfaction. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0038] Figure 1 A flowchart diagram of a method for analyzing user search intentions in a website navigation service system based on artificial intelligence is provided in Example 1 of the present invention.

[0039] Figure 2 A schematic diagram of the module composition of an artificial intelligence-based website navigation service system provided in Embodiment 2 of the present invention. DETAILED DESCRIPTION

[0040] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0041] Embodiment 1: This embodiment can be applied to the scenario of website navigation. The method can be executed by an artificial intelligence-based website navigation service system provided by this embodiment. The method specifically includes the following steps:

[0042] Step S1: extract the term information currently searched by the user on the front-end page, as well as the user's historical search record data and the user's IP address data;

[0043] Step S2: dividing the obtained historical search record data into keywords to obtain historical keywords, storing them in a backend database, processing them according to the searched term information to obtain term keywords, and then searching for related keywords in the system library according to the term keywords;

[0044] Step S3: semantically integrate each keyword with the associated keywords to obtain a long-tail keyword array, and search based on these long-tail keywords;

[0045] Step S4: sorting the search terms according to the results of the system analysis through the search engine evaluation algorithm, and sending the final results to the search page;

[0046] Step S5: The search results are automatically fed back to the backend server, and the user provides feedback on the search results based on the opinion interface built by the backend server.

[0047] In an embodiment of the present invention, the step S2 further includes the following steps:

[0048] Step S21, the search engine performs keyword processing based on the user's historical search record data to obtain the historical keywords. When weighting the historical keywords, the more recent the appearance time, the greater the weight of the historical keyword, and the more frequent the appearance, the greater the weight of the historical keyword. The weight value of the historical keyword division is calculated by the formula:

[0049]

[0050] S f =n i

[0051] S=αS t +(1-α)S f

[0052] Where m represents the time decay factor and decreases with time, t represents the current time, and t i Indicates the time when the keyword last appeared, T represents a time constant, S t represents the time weight, n i Indicates the number of times the keyword appears, S f represents the frequency weight, α represents a parameter for adjusting the importance of time weight and frequency weight, and S represents the final weight of the historical keyword;

[0053] Step S22: After obtaining the historical keyword, the backend database will attach multiple tag indexes to the historical keyword according to the associated scope words of the historical keyword to construct a multi-dimensional historical keyword array;

[0054] Step S23, the search engine divides the searched term information into the term keywords, and uses the PageRank algorithm to evaluate the weight of each term keyword, and divides them into the first term keyword and the second term keyword. The lower the weight of the keyword, the lower the order, and they are identified with numbers according to the order of the word sequence. However, when the position of a term keyword is swapped with another term keyword and the semantics remain unchanged, the above two term keywords are identified with the same number, and an array of term keywords with subscript identifiers will be obtained. Then, according to these keywords, related keywords are retrieved in the system library to obtain a multi-dimensional array of related keywords corresponding to each term keyword.

[0055] In an embodiment of the present invention, the step S23 further includes the following steps:

[0056] Step S231: When the search engine is dividing the entry keywords, if there are more than two entry keywords, the historical keywords analyzed in step S21 are matched with the entry keywords, and the matching value between the entry keywords and the historical keywords is determined by semantic analysis technology. When the matching value reaches a set threshold, the historical keywords are used as the entry keywords of the searched entry information, thereby further narrowing the search scope and improving the accuracy and relevance of the search;

[0057] Step S232: On the contrary, when the search engine is dividing the entry keywords, if only one or two of the entry keywords are obtained, resulting in an unclear search range, step S221 will be repeated to determine whether there are relevant historical keywords that match them;

[0058] Step S233: The search engine attaches multiple tag indexes to the entry keyword according to the associated scope words of the entry keyword, constructs a multidimensional entry keyword array, and calculates the Mahalanobis distance d between the multidimensional entry keyword array and the multidimensional historical keyword array through the multidimensional historical keyword array analyzed in step S22:

[0059]

[0060] In the formula, α represents a multidimensional entry keyword array, β represents a multidimensional historical keyword array, ∑ -1 represents the inverse matrix of the covariance matrix;

[0061] The calculated Mahalanobis distance is compared with the preset value, and the historical keywords corresponding to the distance less than the preset condition are used as the associated keywords of the entry keyword. The historical keywords that meet the conditions are sorted according to the weight of the closer the distance d is, the greater the weight is. By judging the user's historical search records, the scope is narrowed and the accuracy and relevance of the search are improved.

[0062] In an embodiment of the present invention, step S3 further includes the following steps:

[0063] Step S31, the search engine calculates the Mahalanobis distance between the analyzed multidimensional entry keyword array and the multidimensional associated keyword array, and sorts the calculated Mahalanobis distance results in ascending order to obtain a distance sorting table, selects the top-ranked associated keywords, and makes the number of the entry keyword identification consistent with the number to be identified for the corresponding associated keyword;

[0064] Step S32, sequentially and orderly semantically integrate the corresponding entry keywords and the associated keywords in the order of sorting, and perform semantic association between the entry keywords and the associated keywords according to the digital identifier in the integration basis to obtain the long-tail keyword array, and weight sort the long-tail keywords according to the Mahalanobis distance result of step S31, so as to better meet the actual search intention of the user and prevent the occurrence of results contrary to the user's intention;

[0065] Step S33, after obtaining the long-tail keyword array, compare and analyze it with the preset content of the back-end database to determine whether the user IP address data is needed to further divide the weight of the long-tail keyword. If it is determined to be yes, the search engine will dynamically adjust the weight of the long-tail keyword based on the user IP address data, or provide more relevant localized search results.

[0066] In an embodiment of the present invention, step S4 further includes: the search engine ranks the long-tail keywords based on the weights obtained in the analysis, the higher the ranking, the higher the matching degree with the term searched by the user, the closer the long-tail keyword with a high matching degree in the search item, the higher the ranking, and the smaller the corresponding item number.

[0067] In an embodiment of the present invention, step S5 further includes: providing a feedback button or link on the search results page to allow users to evaluate or make suggestions on the search results, automatically triggering optimization of the search analysis algorithm based on the feedback information, and using JavaScript to monitor the user's search behavior, monitor the user's search habits entered on the front-end page, the click-through rate of the search results, the specific items clicked, and the time spent on each search result, using the above data as samples to train and improve the search engine's keyword extraction algorithm and adjust the weights of related keywords to ensure that the search engine can more accurately understand the user's query intent and provide more personalized search results.

[0068] Embodiment 2: Embodiment 2 of the present invention provides a website navigation service system based on artificial intelligence. Figure 2A schematic diagram of the module composition of a website navigation service system based on artificial intelligence provided in the second embodiment of the present invention is shown in FIG. Figure 2 As shown, the system includes a data acquisition module, a keyword extraction module, a keyword aggregation module, a demand matching module and a result feedback module:

[0069] The data acquisition module is used to acquire the information of the term currently searched by the user, as well as the user's historical search record data and the user's IP address data;

[0070] The keyword extraction module is used to extract the information of the term currently searched by the user, the keywords of the user's historical search record data, and the associated keywords;

[0071] The keyword aggregation module is used to semantically integrate the acquired keywords to form long-tail keywords;

[0072] The demand matching module is used to match the user's search demand according to the weight of its long-tail keywords and the user's IP address data;

[0073] The result feedback module is used to monitor the user's search behavior and the feedback data actively provided by the user to adjust the background sample data and automatically update the website information.

[0074] In some embodiments of the present invention, the keyword extraction module includes a multi-dimensional historical keyword module, a multi-dimensional entry keyword module, and a multi-dimensional associated keyword module:

[0075] The multi-dimensional historical keyword module is used to analyze the user's search habits and locate the user's search intention;

[0076] The multi-dimensional keyword module is used to extract keywords from the user's current search terms and divide their keyword ranges to facilitate matching with the historical keyword module;

[0077] The multi-dimensional associated keyword module is used to further expand the keywords of the entry keywords and add associated keywords.

[0078] In some embodiments of the present invention, the keyword aggregation module includes a long-tail keyword module and a demand matching module:

[0079] The long-tail keyword module is used to perform phonetic integration of the keywords analyzed by the keyword extraction module;

[0080] The demand matching module is used to sort the integrated long-tail keywords by the calculated weight values, and match the user search needs in combination with the acquired user IP address data.

[0081] In some embodiments of the present invention, the result feedback module includes a website information automatic update module and a background data maintenance module:

[0082] The website information automatic update module is used to ensure the real-time nature of the website content and provide users with the latest search results;

[0083] The background data maintenance module is used to monitor the user's search behavior and the feedback data actively provided by the user to adjust the background sample data.

[0084] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0085] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An artificial intelligence-based website navigation service system, which executes an artificial intelligence-based website navigation method, characterized in that: The method specifically comprises the following steps: Step S1: extract the term information currently searched by the user on the front-end page, as well as the user's historical search record data and the user's IP address data; Step S2: dividing the obtained historical search record data into keywords to obtain historical keywords, storing them in a backend database, processing them according to the searched term information to obtain term keywords, and then searching for related keywords in the system library according to the term keywords; Step S3: semantically integrate each keyword with the associated keywords to obtain a long-tail keyword array, and search based on these long-tail keywords; Step S4: sorting the search terms according to the results of the system analysis through the search engine evaluation algorithm, and sending the final results to the search page; Step S5: The search results are automatically fed back to the backend server, and the user provides feedback on the search results based on the opinion interface built by the backend server.

2. The website navigation service system based on artificial intelligence according to claim 1, characterized in that: The step S2 further comprises the following steps: Step S21, the search engine performs keyword processing based on the user's historical search record data to obtain the historical keywords. When weighting the historical keywords, the more recent the appearance time, the greater the weight of the historical keyword, and the more frequent the appearance, the greater the weight of the historical keyword. The weight value of the historical keyword division is calculated by the formula: S f =n i S=αS t +(1-α)S f Where m represents the time decay factor and decreases with time, t represents the current time, and t i Indicates the time when the keyword last appeared, T represents a time constant, S t represents the time weight, n i Indicates the number of times the keyword appears, S f represents the frequency weight, α represents a parameter for adjusting the importance of time weight and frequency weight, and S represents the final weight of the historical keyword; Step S22: After obtaining the historical keyword, the backend database will attach multiple tag indexes to the historical keyword according to the associated scope words of the historical keyword to construct a multi-dimensional historical keyword array; Step S23, the search engine divides the searched term information into the term keywords, and uses the PageRank algorithm to evaluate the weight of each term keyword, and divides them into the first term keyword and the second term keyword. The lower the weight of the keyword, the lower the order, and they are identified with numbers according to the order of the word sequence. However, when the position of a term keyword is swapped with another term keyword and the semantics remain unchanged, the above two term keywords are identified with the same number, and an array of term keywords with subscript identifiers will be obtained. Then, according to these keywords, related keywords are retrieved in the system library to obtain a multi-dimensional array of related keywords corresponding to each term keyword.

3. The website navigation service system based on artificial intelligence according to claim 2, characterized in that: The step S23 further comprises the following steps: Step S231: When the search engine divides the entry keyword, if there are more than two entry keywords, the search engine will match the entry keyword with the historical keyword analyzed in step S21, and determine the matching value between the entry keyword and the historical keyword by semantic analysis technology. When the matching value reaches a set threshold, the historical keyword is used as the entry keyword of the searched entry information; Step S232: On the contrary, when the search engine is dividing the entry keywords, if only one or two of the entry keywords are obtained, resulting in an unclear search range, step S221 will be repeated to determine whether there are relevant historical keywords that match them; Step S233: The search engine attaches multiple tag indexes to the entry keyword according to the associated scope words of the entry keyword, constructs a multidimensional entry keyword array, and calculates the Mahalanobis distance d between the multidimensional entry keyword array and the multidimensional historical keyword array through the multidimensional historical keyword array analyzed in step S22: In the formula, α represents a multidimensional entry keyword array, β represents a multidimensional historical keyword array, ∑ -1 represents the inverse matrix of the covariance matrix; The calculated Mahalanobis distance is compared with the preset value, and the historical keywords corresponding to the distance less than the preset condition are used as the associated keywords of the entry keyword, and the historical keywords that meet the conditions are sorted according to the closer the distance d, the greater the weight.

4. The website navigation service system based on artificial intelligence according to claim 3 is characterized by: The step S3 further comprises the following steps: Step S31, the search engine calculates the Mahalanobis distance between the analyzed multidimensional entry keyword array and the multidimensional associated keyword array, and sorts the calculated Mahalanobis distance results in ascending order to obtain a distance sorting table, selects the top-ranked associated keywords, and makes the number of the entry keyword identification consistent with the number to be identified for the corresponding associated keyword; Step S32, sequentially and orderly semantically integrate the corresponding entry keywords and the associated keywords in the order of sorting, perform semantic association between the entry keywords and the associated keywords according to the digital identifier in the integration basis, obtain the long-tail keyword array, and weight sort the long-tail keywords according to the Mahalanobis distance result of step S31; Step S33, after obtaining the long-tail keyword array, compare and analyze it with the preset content of the back-end database to determine whether the user IP address data is needed to further divide the weight of the long-tail keyword. If it is determined to be yes, the search engine will dynamically adjust the weight of the long-tail keyword based on the user IP address data, or provide more relevant localized search results.

5. The website navigation service system based on artificial intelligence according to claim 4, characterized in that: The step S4 further includes: the search engine ranks the long-tail keywords based on the weights obtained by the analysis, the higher the ranking, the higher the matching degree with the term searched by the user, the closer the long-tail keyword with a high matching degree in the search item, the higher the ranking, and the smaller the corresponding item number.

6. The website navigation service system based on artificial intelligence according to claim 5, characterized in that: The step S5 further includes: providing a feedback button or link on the search result page to allow users to evaluate or make suggestions on the search results, automatically triggering optimization of the search analysis algorithm based on the feedback information, and using JavaScript to monitor the user's search behavior, monitor the user's search habits entered on the front-end page, the click-through rate of the search results, the specific items clicked, and the time spent on each search result, and use the above data as samples to train and improve the search engine's keyword extraction algorithm and adjust the weights of related keywords.

7. The website navigation service system based on artificial intelligence according to claim 6, characterized in that: It includes data acquisition module, keyword extraction module, keyword aggregation module, demand matching module and result feedback module: The data acquisition module is used to acquire the information of the term currently searched by the user, as well as the user's historical search record data and the user's IP address data; The keyword extraction module is used to extract the information of the term currently searched by the user, the keywords of the user's historical search record data, and the associated keywords; The keyword aggregation module is used to semantically integrate the acquired keywords to form long-tail keywords; The demand matching module is used to match the user's search demand according to the weight of its long-tail keyword and the user's IP address data.

8. The website navigation service system based on artificial intelligence according to claim 7, characterized in that: The keyword extraction module includes a multidimensional historical keyword module, a multidimensional entry keyword module, and a multidimensional associated keyword module. The multidimensional historical keyword module is used to analyze the user's search habits and locate the user's search intentions. The multidimensional entry keyword module is used to extract keywords from the user's current search entry information and divide its keyword range to facilitate matching with the historical keyword module. The multidimensional associated keyword module is used to further expand the keywords of the entry keywords and add associated keywords.

9. The website navigation service system based on artificial intelligence according to claim 8, characterized in that: The keyword aggregation module includes a long-tail keyword module and a demand matching module. The long-tail keyword module is used to perform phonetic integration on the keywords analyzed by the keyword extraction module. The demand matching module is used to sort the integrated long-tail keywords by calculated weight values ​​and match the user search needs in combination with the acquired user IP address data.

10. The website navigation service system based on artificial intelligence according to claim 9, characterized in that: The result feedback module includes a website information automatic update module and a background data maintenance module. The website information automatic update module is used to ensure the real-time nature of the website content and provide users with the latest search results. The background data maintenance module is used to monitor the user's search behavior and the feedback data actively provided by the user to adjust the background sample data.

Images