SEO optimization method and system based on Internet platform algorithm

Abstract

The invention belongs to the field of SEO optimization, relates to a data analysis technology, is used for solving the problem that only a single output sequence can be given when keyword search is carried out on a user in the prior art, and particularly relates to an SEO optimization method and system based on an internet platform algorithm, and the SEO optimization system comprises a preprocessing module, a composite search module and a sorting optimization module which are connected in sequence; the preprocessing module is used for preprocessing keywords input by a user: after the user inputs the keywords, combining the keywords input by the user with L1 associated words one by one to obtain L1 long-tail words, and sending all the long-tail words to the composite search module; according to the method, the keyword input by the user can be preprocessed, the associated word is marked in combination with the keyword input by the user and the input category, the long-tail word is obtained according to the combination of the associated word and the keyword input by the user, and the user demand is extracted more accurately through the long-tail word and search is carried out.

Description

Technical Field

[0001] The present invention belongs to the field of SEO optimization, involves data analysis technology, and specifically is an SEO optimization method and system based on an Internet platform algorithm. Background Art

[0002] SEO is a method that utilizes the internal rules of search engines to optimize website structures and content, thereby enhancing the natural ranking of websites in search engine results. Its core objective is to improve website visibility, obtain brand benefits, and gain more traffic and market competitive advantages for enterprises or individuals.

[0003] The invention patent with the publication number CN112434240A discloses an SEO intelligent optimization ranking and query system. This SEO optimization system improves the rankings of keywords, images, and pages in search engines, thereby accurately attracting users to enter this website, obtaining free traffic, and enabling the advertisements within the website to generate promotional effects. However, when this SEO optimization system conducts keyword searches for users, it can only give a single output sequence, and the output result has a low matching degree with user needs. At the same time, it is unable to optimize the sorting algorithm in combination with the application status of the output sequence, resulting in the inability to guarantee the SEO optimization effect.

[0004] In view of the above technical problems, this application proposes a solution. Summary of the Invention

[0005] The purpose of the present invention is to provide an SEO optimization method and system based on an Internet platform algorithm, which is used to solve the problem that the existing technology can only give a single output sequence when conducting keyword searches for users. The technical problem that the present invention needs to solve is: how to provide an SEO optimization method and system based on an Internet platform algorithm that can conduct composite search analysis when conducting keyword searches for users.

[0006] The purpose of the present invention can be achieved through the following technical solutions: An SEO optimization system based on an Internet platform algorithm includes a preprocessing module, a composite search module, and a sorting optimization module that are connected in sequence. The preprocessing module is used to preprocess the keywords input by users: after a user inputs keywords, the product category corresponding to the keywords is marked as the input category, the related keywords are marked through the input category, and the keywords input by the user are combined with L1 related keywords one by one to obtain L1 long-tail keywords, and all the long-tail keywords are sent to the composite search module. The composite search module is used to perform composite search analysis on the generated long-tail words: generate an analysis cycle, randomly select a ranking algorithm as the analysis algorithm when performing composite search within the analysis cycle, generate L1 single sequences for the L1 long-tail words through the analysis algorithm, perform primary processing and secondary processing on the single sequences to obtain primary composite sequences and secondary composite sequences; generate a composite derived sequence from the primary composite sequence - the secondary composite sequence, and send the composite derived sequence to the user client; The sorting optimization module is used to perform dynamic optimization analysis on the sorting algorithm.

[0007] Furthermore, the specific process of marking associated words includes: marking products in the user's shopping cart, favorites, and purchased products whose product category is the input category as associated products, forming a key set with the keywords of all associated products, marking the number of elements corresponding to the keywords in the key set as the associated value of the keyword, and marking the L1 keywords with the largest associated value as associated words.

[0008] Furthermore, the ranking algorithms include the BERT algorithm, the Neural Matching algorithm, the Mobile-Friendly algorithm, and the Core Web Vitals algorithm.

[0009] Furthermore, the specific process of performing primary processing on a single sequence includes: marking the number of repeated occurrences of the search result in L1 single sequences as the primary priority value of the search result, marking the search result whose primary priority value is not less than a preset primary priority threshold as a primary associated result, and arranging the primary associated results in descending order of the primary priority value to obtain a primary composite sequence.

[0010] Furthermore, the specific process of performing secondary processing on a single sequence includes: eliminating all primary associated results in the single sequence, then obtaining the click-through rate of the remaining search results in the previous analysis cycle and marking them as independent priority values, and arranging all remaining search results in descending order of independent priority values to obtain a secondary composite sequence.

[0011] Furthermore, the specific process of the sorting optimization module for dynamically optimizing and analyzing the sorting algorithm includes: obtaining the click data and bounce data of the sorting algorithm at the end of the analysis period; marking the ratio of the click data to the bounce data as the optimization value of the sorting algorithm, and arranging the BERT algorithm, Neural Matching algorithm, Mobile-Friendly algorithm, and CoreWeb Vitals algorithm in descending order of the optimization value to obtain an optimization sequence, and sequentially assigning random weights k1, k2, k3, and k4 to the sorting algorithm according to the arrangement order of the optimization sequence, where k1 + k2 + k3 + k4 = 1, and k1 > k2 > k3 > k4; when performing a composite search in the next analysis period, randomly select the sorting algorithm according to the random weights.

[0012] Furthermore, the process of obtaining the click data and bounce data includes: respectively obtaining the click data and bounce data of the BERT algorithm, Neural Matching algorithm, Mobile-Friendly algorithm, and Core Web Vitals algorithm when they are used as analysis algorithms. The click data and bounce data are respectively the number of times the user clicks after the analysis algorithm outputs a composite export sequence and the number of times the user exits the page.

[0013] The SEO optimization method based on the Internet platform algorithm includes the following steps: Step P1: Preprocess the keywords input by the user: After the user inputs keywords, mark the product category corresponding to the keywords as the input category, generate L1 associated words through the input category, and combine the keywords input by the user with the L1 associated words one by one to obtain L1 long-tail words; Step P2: Perform composite search analysis on the generated long-tail words: Generate an analysis period, randomly select a sorting algorithm as the analysis algorithm, generate L1 single sequences for the L1 long-tail words through the analysis algorithm, perform primary processing and secondary processing on the single sequences and generate a composite export sequence; Step P3: Perform dynamic optimization analysis on the sorting algorithm: At the end of the analysis period, obtain the optimization values of different sorting algorithms when they are used as analysis algorithms for composite search analysis, and mark the random weights of the sorting algorithm in the next analysis period according to the optimization values.

[0014] The present invention has the following beneficial effects: 1. Through the preprocessing module, the keywords input by the user can be preprocessed, the associated words can be marked in combination with the keywords input by the user and the input category, and long-tail words can be obtained by combining the associated words with the keywords input by the user, so as to more accurately extract the user's needs and perform searches; 2. The generated long-tail keywords can be analyzed through compound search by the compound search module. The analysis algorithm is marked in a random selection manner, and then a single sequence is generated for the long-tail keywords through the analysis algorithm. The priority of the search results in the single sequence is reordered through primary processing and secondary processing to expand the search scope in the form of long-tail keywords, and then the single sequence is reorganized and sorted to generate an optimized search result by combining user preferences and big data trends. 3. The sorting optimization module can perform dynamic optimization analysis on the sorting algorithm. Optimization values are generated based on the click data and bounce data corresponding to different sorting algorithms as the analysis algorithm, and the random selection weights of the sorting algorithm in the next analysis cycle are marked by the optimization values to dynamically optimize the selection priority of the sorting algorithm. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0016] Figure 1 It is the system block diagram of Embodiment 1 of the present invention; Figure 2 It is the method flow chart of Embodiment 2 of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0017] The following will clearly and completely describe the technical solutions of the present invention in conjunction with the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts fall within the scope of protection of the present invention.

[0018] Embodiment 1: As Figure 1 shown, the SEO optimization system based on the Internet platform algorithm includes a preprocessing module, a compound search module, and a sorting optimization module connected in sequence.

[0019] The preprocessing module is used to preprocess the keywords input by the user: after the user enters the keyword, the product category corresponding to the keyword is marked as the input category, and the products with the input category in the user's shopping cart, favorites and purchased products are marked as related products. The keywords of all related products constitute a key set, and the number of elements corresponding to the keyword in the key set is marked as the keyword's association value. The L1 keywords with the largest association value are marked as associated words. The keyword input by the user is combined with the L1 associated words one by one to obtain L1 long-tail words, and all long-tail words are sent to the compound search module; the keyword input by the user is preprocessed, and the associated words are marked in combination with the keyword input by the user and the input category. The long-tail words are combined according to the associated words and the keyword input by the user to obtain user needs and search more accurately.

[0020] The compound search module is used to perform compound search analysis on the generated long-tail words: generate an analysis cycle, and when performing compound search within the analysis cycle, randomly select a sorting algorithm as the analysis algorithm. The sorting algorithms include BERT algorithm, Neural Matching algorithm, Mobile-Friendly algorithm and Core Web Vitals algorithm. The analysis algorithm generates L1 single sequences for L1 long-tail words, and performs primary processing on the single sequence: the number of repeated occurrences of the search results in the L1 single sequences is marked as the primary priority value of the search results, and the search results with a primary priority value not less than the preset primary priority threshold are marked as primary associated results. The primary associated results are arranged in descending order according to the primary priority value to obtain a primary compound sequence; secondary processing is performed on the single sequence: all primary associated results in the single sequence are eliminated, and then the click-through rate of the remaining search results in the previous analysis cycle is obtained and marked as an independent priority value. All remaining search results are arranged in descending order of independent priority values to obtain a secondary composite sequence; a composite derived sequence is generated from the primary composite sequence-secondary composite sequence, and the composite derived sequence is sent to the user client; a composite search analysis is performed on the generated long-tail words, and the analysis algorithm is marked in a random selection manner, and then a single sequence is generated for the long-tail words through the analysis algorithm, and the search results in the single sequence are re-sorted through primary and secondary processing to expand the search range by forming long-tail words, and then the single sequence is reorganized and sorted, and the optimal search results are generated by combining user preferences and big data tendencies.

[0021] The sorting optimization module is used to perform dynamic optimization analysis on sorting algorithms: at the end of the analysis period, obtain the click data and bounce data when using the BERT algorithm, Neural Matching algorithm, Mobile-Friendly algorithm, and Core Web Vitals algorithm as the analysis algorithms respectively. The click data and bounce data are the number of times users click after the analysis algorithm outputs the composite export sequence and the number of times users bounce out of the page respectively; mark the ratio of the click data to the bounce data as the optimization value of the sorting algorithm, and arrange the BERT algorithm, Neural Matching algorithm, Mobile-Friendly algorithm, and Core Web Vitals algorithm in descending order of the optimization value to obtain an optimization sequence. Allocate random weights k1, k2, k3, and k4 to the sorting algorithms in the order of the optimization sequence, where k1 + k2 + k3 + k4 = 1, and k1 > k2 > k3 > k4; during the composite search in the next analysis period, randomly select the sorting algorithms according to the random weights; perform dynamic optimization analysis on the sorting algorithms, generate optimization values through the click data and bounce data corresponding to different sorting algorithms as the analysis algorithms, and mark the random selection weights of the sorting algorithms in the next analysis period through the optimization values to dynamically optimize the selection priority of the sorting algorithms.

[0022] Embodiment 2: As Figure 2 shown, the SEO optimization method based on the Internet platform algorithm includes the following steps: Step P1: Preprocess the keywords input by the user: After the user inputs keywords, mark the product category corresponding to the keywords as the input category, generate L1 associated words through the input category, and combine the keywords input by the user with the L1 associated words one by one to obtain L1 long-tail words; Step P2: Perform composite search analysis on the generated long-tail words: Generate an analysis period, randomly select a sorting algorithm as the analysis algorithm, generate L1 single sequences for the L1 long-tail words through the analysis algorithm, and perform primary processing and secondary processing on the single sequences and generate a composite export sequence; Step P3: Perform dynamic optimization analysis on the sorting algorithms: At the end of the analysis period, obtain the optimization values when different sorting algorithms are used as the analysis algorithms for composite search analysis, and mark the random weights of the sorting algorithms in the next analysis period according to the optimization values.

[0023] SEO optimization method and system based on Internet platform algorithm. When working, after the user inputs a keyword, the product category corresponding to the keyword is marked as the input category, L1 related words are generated through the input category, and the keyword input by the user is combined with the L1 related words one by one to obtain L1 long-tail words; an analysis period is generated, a sorting algorithm is randomly selected as the analysis algorithm, and L1 single sequences are generated for the L1 long-tail words through the analysis algorithm. The single sequences are subjected to primary processing and secondary processing to generate a composite export sequence; at the end of the analysis period, the optimization values when different sorting algorithms are used as the analysis algorithm for composite search analysis are obtained, and the random weights of the sorting algorithm in the next analysis period are marked according to the optimization values.

[0024] The above content is only an example and illustration of the structure of the present invention. Those skilled in the art of this technology can make various modifications or supplements to the described specific embodiments or use similar methods to replace them. As long as they do not deviate from the structure of the invention or exceed the scope defined by this claim book, they should fall within the protection scope of the present invention.

[0025] In the description of this specification, the descriptions referring to terms such as "one embodiment", "example", "specific example", etc. mean that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in a suitable manner in any one or more embodiments or examples.

[0026] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not elaborate on all the details, nor do they limit the invention to only the specific implementation manners. Obviously, according to the content of this specification, many modifications and changes can be made. This specification selects and specifically describes these embodiments to better explain the principle and practical application of the present invention, so that those skilled in the art of this technology can well understand and utilize the present invention. The present invention is only limited by the claim book and its full scope and equivalents.

Claims

1. An SEO optimization system based on Internet platform algorithms, characterized in that, It includes a preprocessing module, a composite search module, and a sorting optimization module that are connected in sequence; The preprocessing module is used to preprocess the keywords input by the user: after the user inputs keywords, the product category corresponding to the keywords is marked as the input category, the related words are marked through the input category, the keywords input by the user are combined with L1 related words one by one to obtain L1 long-tail words, and all the long-tail words are sent to the composite search module; The composite search module is used to perform composite search analysis on the generated long-tail words: generate an analysis period, when performing composite search within the analysis period, randomly select a sorting algorithm as the analysis algorithm, generate L1 single sequences for the L1 long-tail words through the analysis algorithm, and perform primary processing and secondary processing on the single sequences to obtain a primary composite sequence and a secondary composite sequence; generate a composite export sequence from the primary composite sequence - secondary composite sequence, and send the composite export sequence to the user client; The sorting optimization module is used to perform dynamic optimization analysis on the sorting algorithm.

2. The SEO optimization system based on the algorithm of the Internet platform according to claim 1, characterized in that The specific process of marking related words includes: marking the products with the product category of the input category in the user's shopping cart, favorites, and purchased products as related products, forming a key set from the keywords of all related products, marking the number of elements corresponding to the keywords in the key set as the association value of the keywords, and marking the L1 keywords with the largest association value as related words.

3. The SEO optimization system based on the algorithm of the Internet platform according to claim 2, wherein The sorting algorithms include the BERT algorithm, the Neural Matching algorithm, the Mobile-Friendly algorithm, and the Core Web Vitals algorithm.

4. The SEO optimization system based on the algorithm of the Internet platform according to claim 3, characterized in that, The specific process of performing primary processing on the single sequence includes: marking the number of repeated occurrences of the search results in the L1 single sequences as the primary priority value of the search results, marking the search results with the primary priority value not less than the preset primary priority threshold as the primary associated results, and arranging the primary associated results in descending order of the primary priority value to obtain a primary composite sequence.

5. The SEO optimization system based on the algorithm of the Internet platform according to claim 4, characterized in that The specific process of performing secondary processing on the single sequence includes: removing all the primary associated results in the single sequence, then obtaining the click-through rate of the remaining search results in the previous analysis period and marking it as the independent priority value, and arranging all the remaining search results in descending order of the independent priority value to obtain a secondary composite sequence.

6. The SEO optimization system based on the algorithm of the Internet platform according to claim 5, characterized in that The specific process of the sorting optimization module for dynamically optimizing and analyzing sorting algorithms includes: obtaining the click data and bounce data of the sorting algorithm at the end of the analysis period; marking the ratio of the click data to the bounce data as the optimization value of the sorting algorithm, and arranging the BERT algorithm, Neural Matching algorithm, Mobile-Friendly algorithm, and Core Web Vitals algorithm in descending order of the optimization value to obtain an optimization sequence, and sequentially assigning random weights k1, k2, k3, and k4 to the sorting algorithms according to the arrangement order of the optimization sequence, where k1 + k2 + k3 + k4 = 1, and k1 > k2 > k3 > k4; when performing a composite search in the next analysis period, randomly select the sorting algorithms according to the random weights.

7. The SEO optimization system based on the Internet platform algorithm according to claim 6, characterized in that, The process of obtaining the click data and bounce data includes: respectively obtaining the click data and bounce data of the BERT algorithm, Neural Matching algorithm, Mobile-Friendly algorithm, and Core Web Vitals algorithm when they are used as analysis algorithms. The click data and bounce data are respectively the number of times the user clicks after the analysis algorithm outputs a composite export sequence and the number of times the user bounces from the page.

8. An SEO optimization method based on the algorithm of an Internet platform, characterized in that, It includes the following steps: Step P1: Preprocess the keywords input by the user: After the user inputs keywords, mark the product category corresponding to the keywords as the input category, generate L1 related keywords through the input category, and combine the keywords input by the user with the L1 related keywords one by one to obtain L1 long-tail keywords; Step P2: Conduct a composite search analysis on the generated long-tail keywords: Generate an analysis period, randomly select a sorting algorithm as the analysis algorithm, generate L1 single sequences for the L1 long-tail keywords through the analysis algorithm, perform primary processing and secondary processing on the single sequences and generate a composite export sequence; Step P3: Conduct a dynamic optimization analysis on the sorting algorithm: At the end of the analysis period, obtain the optimization values of different sorting algorithms when they are used as analysis algorithms for composite search analysis, and mark the random weights of the sorting algorithms in the next analysis period according to the optimization values.

Images