E-commerce platform online advertisement competitive product recommendation method

Abstract

The invention belongs to the technical field of pushing algorithms, and discloses an e-commerce platform online advertisement competitive product recommendation method, which comprises the steps of capturing commodity information data, performing data cleaning on the commodity information data, performing model construction by using the cleaned data, analyzing consumer preferences by using a clustering and classification method, predicting a market trend, and performing online advertisement competitive product recommendation on the basis of the constructed model. And competitive products are recommended by using machine learning and an artificial intelligence algorithm. According to the scheme, the commodity search result and the detail page information are automatically captured in real time, the difficulty of manually selecting competitive products is reduced, the competitive products are recommended through an intelligent algorithm, the commodity orientation accuracy of advertisements is improved, and the purposes of optimizing the advertisement putting effect and promoting sales are achieved.

Description

Technical Field

[0001] The present application relates to the technical field of push algorithms, and more specifically, to a method for recommending competitive products for online advertising on an e-commerce platform. Background Art

[0002] Amazon is a comprehensive online marketplace that allows sellers to sell products on its platform. Product targeting is an advertising strategy that merchants use on a daily basis. This form of advertising helps to increase product exposure and click-through rate, increase sales, and provides sellers with an effective promotion method to attract more users to buy their products.

[0003] The capabilities currently provided by Amazon include: product classification and refinement. Merchants can filter products based on industry categories and brands, and can also filter based on price and star ratings to target a batch of products; Amazon recommends similar and related products to merchants;

[0004] Product targeting advertising technology on Amazon's e-commerce platform is an advanced marketing strategy that allows sellers to optimize their advertising and market positioning by analyzing and researching competitor products. This technology usually involves the following key steps:

[0005] (1) Competitive product analysis: Sellers first need to identify competitors in the market and conduct in-depth analysis of these competing products. This includes understanding the price, quality, function, user evaluation, and market performance of competing products;

[0006] (2) Keyword research: Using tools, sellers can discover potential popular keywords and understand the keyword rankings of competitors. This helps optimize product titles, descriptions, and background keywords, and improve product rankings in search results;

[0007] (3) Competitive product monitoring: Through the competitive product monitoring function, sellers can track the performance and ranking changes of competitors' products. This helps to analyze competitors' strategies and optimize their own product positioning and promotion strategies accordingly;

[0008] (4) Advertising keyword optimization: Sellers can use tools to identify the most promising advertising keywords and optimize advertising delivery based on keyword analysis to increase the click-through rate and conversion rate of ads.

[0009] (5) Advertising effect monitoring: real-time monitoring of advertising performance and cost-effectiveness, and adjustment of advertising budget and delivery strategy based on monitoring results to maximize advertising effect;

[0010] (6) Competitor advertising analysis: monitor competitors’ advertising and strategies, analyze competitors’ performance, and develop targeted advertising counterattack strategies to seize market share;

[0011] (7) Display Advertising: Amazon display advertising provides sellers with the opportunity to advertise on third-party websites to customers who have viewed their products on Amazon. Such ads can appear on the Amazon homepage, product detail pages, and search results pages, and can also be displayed on third-party websites and applications outside of Amazon;

[0012] (8) Targeted advertising settings: Sellers can filter out competing ASINs and create targeted advertising campaigns by setting different preconditions, such as price, number of reviews, ratings, and category BSR, to accurately target potential customers;

[0013] (9) Contextual targeting: Amazon’s display promotion product targeting has evolved into contextual targeting, which means that advertisers can place relevant ads when customers view pages related to the targeted products, thereby increasing the relevance and attractiveness of the ads.

[0014] Through these technologies and strategies, Amazon sellers can promote their products more effectively, increase exposure and sales, and gain an advantage in the fierce market competition.

[0015] When Amazon sellers manually locate and analyze competing products, they usually take the following steps and may face some challenges and shortcomings.

[0016] (1) Market research: Sellers usually collect market data through manual search and recording, which includes information such as prices, sales volume, and user reviews of competing products. However, this method is inefficient.

[0017] (2) Competitive product selection: Sellers identify competitors based on market performance and product similarity, but this process often relies on intuition and experience and lacks data support.

[0018] (3) Data collection: Manually collecting competitive product information, such as product lists, promotional activities, etc. This process is tedious and prone to errors.

[0019] (4) SWOT analysis: Conducting a SWOT analysis requires a deep understanding of the market and products, but manual analysis may be limited due to a lack of comprehensive data.

[0020] (5) Keyword analysis: Sellers manually search and record competitor keywords, but this usually does not include in-depth keyword trend analysis.

[0021] (6) User review analysis: Manually reading and analyzing user reviews is time-consuming and may affect the analysis results due to personal bias.

[0022] (7) Sales strategy analysis: Researching the sales strategies of competing products requires an in-depth understanding of market dynamics, but manual analysis may not be able to fully cover all strategies.

[0023] (8) Product positioning: Determining product positioning based on limited data may lead to inaccurate positioning.

[0024] (9) Strategy formulation: When formulating market strategies, the lack of real-time data support may lead to inflexible strategies.

[0025] In summary, existing technologies rely on manual positioning and analysis of competitors, which requires a lot of time, which may affect sellers' attention to other important tasks. Secondly, manually collected data may miss key information, such as competitors' inventory levels, supply chain issues, etc., and the analysis results may be biased due to personal experience and subjective judgment. For small sellers, it may be impractical to invest a lot of human resources in competitor analysis. Due to the lack of advanced analytical tools and algorithm support, it is difficult to conduct in-depth data mining and predictive analysis, and it is difficult to obtain internal data of competitors, such as cost structure, supply chain information, etc., which are crucial for formulating strategies. In addition, the manual analysis and decision-making process is slow and cannot be updated in real time, resulting in the inability to capture market dynamics in a timely manner, and thus missing market opportunities. As the market size expands, human resources need to be continuously invested, and the sustainability of manual analysis has become an issue that needs to be urgently addressed. Summary of the invention

[0026] In order to solve the above problems, the present application provides an e-commerce platform online advertising competitive product recommendation method, which aims to optimize the efficiency of sellers on the Amazon e-commerce platform in selecting and analyzing competitive products.

[0027] The present application provides an e-commerce platform online advertising competitive product recommendation method using the following technical solutions:

[0028] A method for recommending competitive products for online advertising on an e-commerce platform, comprising the following steps:

[0029] S1. Use a web crawler to automatically crawl product information data from the Amazon e-commerce platform, including but not limited to product ASIN, detail page content, user reviews, price changes, and sales rankings;

[0030] S2. Clean the product information data, remove invalid, erroneous and incomplete data, and unify the data format;

[0031] S3. Use the cleaned data to build models, use clustering and classification methods to analyze consumer preferences and predict market trends;

[0032] S4. Based on the constructed model, use machine learning and artificial intelligence algorithms to recommend competing products.

[0033] Furthermore, the data cleaning includes: identifying and deleting duplicate data entries; converting data from different sources into a unified format; identifying and processing abnormal values ​​and outliers; and filling or deleting missing data.

[0034] Furthermore, the model construction includes the following steps:

[0035] S301. Find the ASIN of the product, the ASIN of the competing product, and their association in the processed product information database;

[0036] S302, competing products exclude ASINs in this Profile;

[0037] S303, obtain the ASIN details page data of competing products;

[0038] S304: Determine whether the last level category in the block of the competitor's ASIN is consistent with the ASIN of the product. If they are consistent, mark the competitor's ASIN as a similar product. If they are inconsistent, mark the competitor's ASIN as a related product.

[0039] S305: Group similar products according to their associations, and perform data statistics on the groups to obtain brand distribution, price distribution, star rating distribution, and rating quantity distribution.

[0040] Furthermore, the grouping includes: all competing product ASINs, product details pages, keyword searches, BSRs (Amazon Best Sellers List), Categories (category searches), and NNRs (Amazon New Products List).

[0041] Furthermore, obtaining the ASIN detail page data of the competitor product includes: obtaining the block in the ASIN Listing page of the product; using the last level category in the block to find the corresponding URL in the am_rank_index table; if the URL cannot be found in the last level category of the block, use the second to last level and fetch upwards in sequence; and crawling the corresponding ASIN detail page data according to the obtained URL.

[0042] Furthermore, the machine learning and intelligent algorithms include: calculating and evaluating the similarities between products; providing personalized competitive product recommendations based on the seller's product characteristics and market positioning, and responding to market changes and user needs in real time; and self-learning and optimization based on user feedback and recommendation effects.

[0043] In summary, the present application includes at least one of the following beneficial technical effects:

[0044] This solution reduces the difficulty of manually selecting competing products by automatically crawling product search results and detail page information in real time, and recommends competing products through intelligent algorithms, thereby improving the accuracy of product targeting in advertisements, thereby achieving the goal of optimizing advertising effectiveness and promoting sales. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 A schematic diagram of the steps of this application;

[0046] Figure 2 A schematic diagram of the steps for model construction;

[0047] Figure 3 This is a statistical diagram of grouped data. DETAILED DESCRIPTION

[0048] The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application; obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments, and all other embodiments obtained by ordinary technicians in this field based on the embodiments in the present application without making creative work are within the scope of protection of the present application.

[0049] Embodiment 1:

[0050] The present application is further described in detail below in conjunction with the accompanying drawings.

[0051] The present application embodiment discloses a method for recommending competitive products on an online advertisement of an e-commerce platform, comprising the following steps:

[0052] S1. Use a web crawler to automatically crawl product information data from the Amazon e-commerce platform, including but not limited to product ASIN, detail page content, user reviews, price changes, and sales rankings;

[0053] S2. Clean the product information data, remove invalid, erroneous and incomplete data, and unify the data format;

[0054] S3. Use the cleaned data to build models, use clustering and classification methods to analyze consumer preferences and predict market trends;

[0055] S4. Based on the constructed model, use machine learning and artificial intelligence algorithms to recommend competing products.

[0056] like Figure 2 As shown, model construction includes the following steps:

[0057] S301. Find the ASIN of the product, the ASIN of the competing product, and their association in the processed product information database;

[0058] S302, competing products exclude ASINs in this Profile;

[0059] S303, obtain the ASIN details page data of competing products;

[0060] S304: Determine whether the last level category in the block of the competitor's ASIN is consistent with the ASIN of the product. If they are consistent, mark the competitor's ASIN as a similar product. If they are inconsistent, mark the competitor's ASIN as a related product.

[0061] S305, group similar products according to their associations, including all competing product ASINs, product details pages, keyword searches, BSRs (Amazon Best Sellers List), Categories (category searches), and NNRs (Amazon New Products List). Figure 3 As shown, the groups are statistically analyzed to obtain brand distribution, price distribution, star rating distribution, and rating quantity distribution.

[0062] The process of obtaining the ASIN detail page data of competing products includes: obtaining the block in the ASIN Listing page of the product; using the last-level category in the block to find the corresponding URL in the am_rank_index table; if the URL cannot be found in the last-level category of the block, use the second-to-last level and go up in sequence; and crawl the corresponding ASIN detail page data according to the obtained URL.

[0063] The implementation principle of the online advertising competitive product recommendation method of an e-commerce platform in the present application embodiment is as follows:

[0064] The proposal of this application can be broken down into the following four key aspects:

[0065] 1. Crawling

[0066] Crawling is the data collection phase of the entire technical solution. Develop advanced web crawlers to automatically crawl product information from the Amazon e-commerce platform, including but not limited to product search results, detail page content, user reviews, price changes, sales rankings, etc.

[0067] Its technical points include: Multi-threaded crawling: In order to improve efficiency, the crawler uses multi-threaded technology to crawl multiple pages at the same time. Dynamic content processing: It can process content dynamically generated by JavaScript to ensure the integrity of the data. Anti-crawler strategy: Design strategies to circumvent Amazon's anti-crawler mechanism to ensure the continuity and stability of data crawling. Data update frequency: Set a reasonable data update frequency to ensure the timeliness of information.

[0068] 2. Data cleaning

[0069] Data cleaning is a step to ensure data quality. It involves removing invalid, erroneous or incomplete data, and unifying the data format to lay a solid foundation for subsequent analysis and modeling.

[0070] Its technical points include: Deduplication: Identify and delete duplicate data entries. Format standardization: Convert data from different sources into a unified format. Outlier detection: Identify and handle outliers or outliers. Missing value processing: Determine the strategy for filling or deleting missing data.

[0071] 3. Data Modeling

[0072] Data modeling is the use of cleaned data to build models that can analyze market trends, consumer behavior, and product performance.

[0073] Its technical points include: Feature engineering: extracting features from raw data that help the model understand. Market trend analysis: building models to predict market trends and consumer behavior. Consumer behavior analysis: using clustering, classification and other methods to analyze consumer preferences. Product performance prediction: building a prediction model to estimate the future market performance of the product.

[0074] 4. Intelligent Recommendation

[0075] Intelligent recommendation is the core of the entire technical solution. It uses machine learning and artificial intelligence algorithms to recommend competing products based on the output of the previous steps.

[0076] The technical points include: developing algorithms to calculate and evaluate the similarities between products. Providing personalized competitive product recommendations based on the seller's product characteristics and market positioning. Real-time recommendations to ensure that the recommendation system can respond to market changes and user needs in real time. The system can self-learn and optimize based on user feedback and recommendation results.

[0077] Through close collaboration in these four stages, this technical solution can provide Amazon sellers with a comprehensive, efficient and intelligent competitive product analysis and recommendation service, helping them optimize their advertising strategies and improve sales performance.

[0078] Using the above patented technology, Amazon sellers can expect the following results:

[0079] 1. Improve efficiency

[0080] Through automated crawler technology, sellers can quickly obtain a large amount of competitive product data, saving time on manual search and data collation. The automated process reduces the possibility of human input errors and improves the accuracy of data processing.

[0081] 2. Improve data quality and depth

[0082] The crawler can cover a wide range of data points, including dynamic information that is difficult to track manually, such as real-time price and inventory changes. Data cleaning and modeling ensure the depth and quality of analysis, helping sellers gain deeper market insights.

[0083] 3. More precise market positioning

[0084] Intelligent algorithms recommend competing products based on product features and market performance, helping sellers accurately identify market opportunities. Based on the intelligently recommended competitive product information, sellers can develop more personalized and targeted advertising strategies.

[0085] 4. Enhanced decision support

[0086] Sellers can make smarter business decisions based on high-quality data and analysis results. Real-time data updates and intelligent recommendations enable sellers to quickly adapt to market changes.

[0087] 5. Optimized advertising delivery

[0088] The intelligent recommendation system ensures that the advertising content is highly relevant to the interests and needs of target consumers. Through precise market positioning and personalized advertising strategies, the return on advertising investment is improved.

[0089] 6. Boost sales and growth

[0090] An optimized advertising strategy can increase product exposure to potential buyers. Through precise market positioning, sellers can attract more consumers with purchasing intentions, thereby increasing conversion rates.

[0091] 7. Competitive Advantage

[0092] By quickly acquiring and analyzing market data, sellers are able to stay ahead of the competition. The system's self-learning and optimization capabilities ensure long-term competitive advantage.

[0093] 8. Risk Management

[0094] Real-time data updates enable sellers to respond to market changes in a timely manner and reduce market risks. The intelligent recommendation system provides the ability to flexibly adjust strategies to cope with changing market conditions.

[0095] In summary, this patented technology provides Amazon sellers with a powerful tool to improve operational efficiency, optimize market strategies, and ultimately drive sales growth and brand development.

[0096] The above are all preferred embodiments of the present application, and the protection scope of the present application is not limited thereto. Therefore, any equivalent changes made according to the structure, shape, and principle of the present application should be included in the protection scope of the present application.

Claims

1. A method for recommending competitive products for online advertising on an e-commerce platform, characterized in that: The following steps are involved: S1. Use a web crawler to automatically crawl product information data from the Amazon e-commerce platform, including but not limited to product ASIN, detail page content, user reviews, price changes, and sales rankings; S2. Clean the product information data, remove invalid, erroneous and incomplete data, and unify the data format; S3. Use the cleaned data to build models, use clustering and classification methods to analyze consumer preferences and predict market trends; S4. Based on the constructed model, use machine learning and artificial intelligence algorithms to recommend competing products.

2. The method for recommending competitive products for online advertising on an e-commerce platform according to claim 1, characterized in that: The data cleaning includes: Identify and remove duplicate data entries; convert data from different sources into a unified format; identify and handle abnormal values ​​and outliers; fill in or delete missing data.

3. The method for recommending competitive products for online advertising on an e-commerce platform according to claim 1, characterized in that: The model construction includes the following steps: S301. Find the ASIN of the product, the ASIN of the competing product, and their association in the processed product information database; S302, competing products exclude ASINs in this Profile; S303, obtain the ASIN details page data of competing products; S304: Determine whether the last level category in the block of the competitor's ASIN is consistent with the ASIN of the product. If they are consistent, mark the competitor's ASIN as a similar product. If they are inconsistent, mark the competitor's ASIN as a related product. S305: Group similar products according to their associations, and perform data statistics on the groups to obtain brand distribution, price distribution, star rating distribution, and rating quantity distribution.

4. The method for recommending competitive products for online advertising on an e-commerce platform according to claim 3, characterized in that ,The groups include: all competing ASINs, product details pages, keyword searches, BSRs (Amazon best sellers list), Categories (category search) and NNRs (Amazon new product list).

5. The method for recommending competitive products for online advertising on an e-commerce platform according to claim 4, characterized in that: Obtaining competitor ASIN detail page data includes: Get the block in the ASIN Listing page of this product; use the last category in the block to find the corresponding URL in the am_rank_index table; if the URL cannot be found in the last category of the block, use the second-to-last category and go up in sequence; crawl the corresponding ASIN detail page data based on the obtained URL.

6. The method for recommending competitive products for online advertising on an e-commerce platform according to claim 1, characterized in that: The machine learning and intelligent algorithms include: Calculate and evaluate the similarities between products; provide personalized competitive product recommendations based on the seller's product features and market positioning, and respond to market changes and user needs in real time; conduct self-learning and optimization based on user feedback and recommendation results.

Images