TW201935363A - Recommended system and method of product promotion combination - Google Patents
Recommended system and method of product promotion combination Download PDFInfo
- Publication number
- TW201935363A TW201935363A TW107103633A TW107103633A TW201935363A TW 201935363 A TW201935363 A TW 201935363A TW 107103633 A TW107103633 A TW 107103633A TW 107103633 A TW107103633 A TW 107103633A TW 201935363 A TW201935363 A TW 201935363A
- Authority
- TW
- Taiwan
- Prior art keywords
- product
- offers
- time period
- value
- discount
- Prior art date
Links
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
Description
本發明關於一種產品優惠組合推薦技術,特別是一種結合數據分析的電信產品優惠組合推薦系統與方法。 The invention relates to a product preference combination recommendation technology, in particular to a telecommunication product preference combination recommendation system and method combined with data analysis.
傳統電信產品進行優惠促銷時,其常見的處理模式通常是採取特定的電信產品的牌價降價優惠,如行動上網商品原始牌價打折促銷,或是多種電信產品的合購具有某種折扣的推薦方式,像是電信產品A與產品B同時訂購則可享較多折扣價格。 When traditional telecommunications products are offered for promotion, the common processing mode is usually to take a discount on the price of specific telecommunications products, such as a discount on the original listing price of mobile Internet products, or a recommended way of combining certain telecommunications products with a certain discount. For example, telecommunication products A and B can be ordered at the same time to enjoy more discounted prices.
然而電信產品及其規格類型多元,所關聯的優惠項目眾多,難以將所有的組合全數列出,故傳統式的產品推薦模式多是以電信產品提供商的業務推廣計劃或是市場調查為基礎,提供幾項重點產品及其相關優惠組合作為行銷策略,然而此種方式卻很可能忽略了消費者真正所感興趣的電信產品,亦無法從消費者所感興趣的電信產品中,推薦合適的優惠組合供其選擇。 However, there are many types of telecommunication products and their specifications, and there are many related preferential items. It is difficult to list all the combinations. Therefore, traditional product recommendation models are mostly based on the business promotion plans or market surveys of telecommunication product providers. Provide several key products and their related preferential combinations as a marketing strategy. However, this method is likely to ignore the telecommunications products that consumers are really interested in, and it is not possible to recommend suitable preferential combinations for telecommunication products that consumers are interested in. Its choice.
換言之,傳統固定折扣或產品合購折扣與消費者預期選用之電信產品之間的關聯性難以衡量,故對於電信公司 而言,此種基於傳統產品推薦方式的行銷宣傳,即便事前投入大量活動行銷經費,但其成效往往需事後才能檢視其影響力,且後續亦難從中分析何種優惠對吸引消費者的注意力能有較顯著之成效。 In other words, the correlation between traditional fixed discounts or product syndication discounts and the telecommunications products that consumers expect to choose is difficult to measure. In terms of this type of marketing promotion based on traditional product recommendation methods, even if a large amount of marketing expenses is invested in advance, its effectiveness often needs to be checked afterwards for its effectiveness, and it is difficult to analyze what kind of discounts will attract consumers ’attention later. Can have more significant results.
此外,對於電信公司而言,由於傳統上習慣從產品的觀點切入,即使能夠有效地經由廣告文宣、影視媒體、網路社群等行銷傳播通路向消費者傳達了產品促銷資訊,但卻難以從中得知何項優惠內容對消費者而言較具吸引力,例如申辦光纖上網,其優惠組合可選擇月租費折抵50元或贈送免費戶外熱點,然而多數消費者可能傾向選擇月租費折抵而非贈送價值更高的免費戶外熱點。 In addition, for telecommunications companies, because they are traditionally used to cut in from the point of view of products, even if they can effectively communicate product promotion information to consumers through marketing channels such as advertising announcements, film and television media, and online communities, it is difficult to learn from them. Learned which preferential content is more attractive to consumers. For example, when applying for fiber optic Internet, its preferential combination can choose to discount the monthly fee of 50 yuan or give away free outdoor hotspots. However, most consumers may prefer to choose a monthly fee discount. Instead of giving away free outdoor hotspots of higher value.
故對於消費者來說,所關心的不外乎是在於如何能夠以最划算的方式獲取如費用折扣減少支出、取得實用的贈品、或是享用優質的加值服務等優惠,也就是說,除了所需要的電信產品之外,是否能有一種方法能快速、即時且貼近消費趨勢的優惠組合推薦方法,協助消費者從電信公司所推出的眾多產品優惠組合方案中,找尋享有高性價比之優惠方案。 Therefore, for consumers, the concern is nothing more than how to get the most cost-effective ways to obtain discounts such as cost discounts, reduce expenses, obtain practical gifts, or enjoy premium value-added services. In other words, in addition to In addition to the required telecommunications products, is there a way to recommend a combination of offers that is fast, instant, and close to consumer trends, helping consumers to find cost-effective offers from the many product portfolios offered by telecommunications companies? .
因此如何改進傳統優惠推薦方式,能夠更加順應時下趨勢潮流、準確並有效吸引消費者目光,便是現行推薦技術亟欲思索突破的關鍵,例如,一先前技術曾提出事先建立大量的商品優惠規則,再根據消費者購物清單之商品與優惠規則進行匹配,快速顯示出該購物清單中一或多個商品可與某一或多個加購商品組合後能享有之折扣金額,雖 可更細膩的依據挑選商品進行推薦,但其依然侷限於供貨廠商所提供之單向商品優惠,所提供之商品優惠與消費者之間並無關聯性,對於消費者購買意願的提升效果有限。另一先前技術曾提出記錄過往已交易之客戶購物清單用以建立關聯規則,當消費者挑選一或多項商品時,透過規則檢索,列出消費者可能會感興趣之一或多個商品,其符合時下大數據分析趨勢,找出消費者潛在感興趣之商品,但卻未考量所選購產品與產品間是否有合購之優惠,或是進一步列出選購產品與現行優惠之搭配方案,提供消費者更高性價比之優質的選擇。 Therefore, how to improve the traditional preferential recommendation method, can be more in line with the current trend, accurately and effectively attract consumers ’attention, is the key to the current recommendation technology eager to think about breakthroughs, for example, a previous technology has proposed to establish a large number of product preferential rules in advance , And then match the product with the preferential rules of the consumer shopping list to quickly display the discount amount that one or more products in the shopping list can be combined with one or more additional purchase products, although Products can be recommended based on more detailed selection, but it is still limited to one-way product offers provided by suppliers. There is no correlation between the offered product offers and consumers, and it has limited effect on improving consumers' willingness to buy. . Another previous technology has proposed to record the previously purchased customer shopping list to establish association rules. When a consumer selects one or more products, the rules are retrieved to list one or more products that the consumer may be interested in. In line with the current big data analysis trend, find out the products that consumers are potentially interested in, but have not considered whether there is a joint purchase offer between the purchased product and the product, or further list the matching plan between the purchased product and the current offer , To provide consumers with a more cost-effective and high-quality choice.
有鑒於上述所提及之現行推薦技術不足以滿足消費者多元化需求,本案發明人乃亟思改進之法,致力克服傳統上之電信產品單向優惠推薦之行銷策略中難以有效推薦等困難,以優惠內容為出發點,導入數據分析技術,提供結合消費趨勢與最佳利益取向的產品優惠組合推薦策略,滿足各種不同消費者族群的購物喜好,精準命中潛在客戶所感興趣之產品標的。 In view of the fact that the current recommendation technology mentioned above is insufficient to meet the diversified needs of consumers, the inventor of this case is eager to improve and strive to overcome difficulties such as the difficulty of effective recommendation in the traditional one-way preferential recommendation marketing strategy for telecommunications products. Based on the preferential content, the introduction of data analysis technology, to provide a combination of product trends and best interests oriented product preference combination recommendation strategy, to meet the shopping preferences of various consumer groups, to accurately target the products of interest to potential customers.
本發明之目的即在於提供一種產品優惠組合推薦系統與方法,結合了產品及優惠的關聯規則,搭配已取用優惠使用量統計及優惠內容價值等資訊,導入數據分析技術以發掘潛在商業脈絡,藉此能確切掌握消費趨勢脈動,提供適切之電信產品優惠組合,讓客戶能以划算的價格取得心目中所想要的電信產品及優惠,創造雙贏。 The purpose of the present invention is to provide a product preference combination recommendation system and method, which combines the association rules of products and offers, and uses information such as the statistics of used preferences and the value of preferential content, and introduces data analysis technology to explore potential business contexts. In this way, we can accurately grasp the pulse of consumer trends and provide appropriate telecommunication product preferential combinations, so that customers can get the telecommunications products and benefits they want at a reasonable price, creating a win-win situation.
可達成上述發明目的之產品優惠組合推薦方法,其包括:取得使用者感興趣的產品;映射該產品以取得該產品的多項優惠,並取得各該優惠的時間週期區間、區間使用量以及優惠價值;計算各該時間週期區間的時間週期權重值;計算各該區間使用量以及與其匹配的該時間週期權重值的乘積的加總後,再除以總時間週期區間以計算得到各該優惠的加權平均使用量;取得各該優惠的內容以量化各該優惠的內容;以及計算各該產品的該多項優惠中各該優惠的該加權平均使用量與其匹配的該優惠價值的乘積後,排序各該產品的該多項優惠的該乘積以作為推薦順序。 A method for recommending a product combination that can achieve the above-mentioned purpose of the invention includes: obtaining a product that the user is interested in; mapping the product to obtain multiple benefits of the product, and obtaining a time period interval, an interval usage amount, and a discount value of each of the benefits ; Calculate the time period weight value of each time period interval; calculate the sum of the product of the usage amount of each interval and the time period weight value matching it, and then divide by the total time period interval to calculate the weight of each discount Average usage; obtain the content of each of the offers to quantify the content of each of the offers; and calculate the product of the weighted average usage of each of the offers in the multiple offers of the product and the value of the offer that matches it, sort each The product of the multiple offers of the product is used as the recommendation order.
本發明另提供一種產品優惠組合推薦系統,其包括:產品資訊處理模組,係用於取得使用者感興趣的產品;產品優惠映射模組,係映射該產品以取得該產品的多項優惠,並取得各優惠的時間週期區間、區間使用量以及優惠價值;優惠使用統計模組,係用以計算各該時間週期區間的時間週期權重值,及計算各該區間使用量以及與其匹配的該時間週期權重值的乘積的加總後,再除以總時間週期區間以計算得到各該優惠的加權平均使用量;優惠價值量化模組,係用於取得各該優惠的內容以量化各該優惠的內容;以及優惠組合推薦模組,係用以計算各該產品的該多項優惠中各該優惠的該加權平均使用量與其匹配的該優惠價值的乘積後,排序各該產品的該多項優惠的該乘積以作為推薦順序。 The present invention further provides a product preference combination recommendation system, which includes: a product information processing module for obtaining a product that is of interest to a user; a product preference mapping module for mapping the product to obtain multiple benefits of the product, and Get the time period interval, interval usage, and discount value of each discount; the discount usage statistics module is used to calculate the time period weight value of each time interval interval, and calculate the usage amount of each interval and the time period matching it After summing up the product of weight values, it is divided by the total time period interval to calculate the weighted average usage of each discount. The discount value quantization module is used to obtain the content of each discount to quantify the content of each discount. ; And the preferential combination recommendation module, which is used to calculate the product of the weighted average usage of each of the multiple offers of the product and the preferential value that matches the product, and then sort the product of the multiple offers of each product As a recommended order.
在前述之產品優惠組合推薦系統與方法中,多項該產 品係集成產品類型,在相同產品類型中多個產品的推薦順序係透過該優惠組合推薦模組以計算各該產品的多個該優惠的各該加權平均使用量與其匹配的該優惠價值的乘積,其加總的結果再除以各該產品中的該優惠的數量並予以排序。 In the aforementioned product preferential combination recommendation system and method, many of the products Strain integrated product type. The recommendation order of multiple products in the same product type is calculated through the product combination recommendation module to calculate the product of each of the weighted average usage of multiple of the product for the product and the value of the product that matches it. , And the total result is divided by the number of the offers in each product and sorted.
在前述之產品優惠組合推薦系統與方法中,該計算各該時間週期區間的時間週期權重值係由下列公式計算得出:W n =1+(n-1)r,其中,Wn表示第n個時間週期區間時的優惠其匹配的區間使用量所分配到的權重,n表示優惠啟用後第n個時間週期區間,r表示可調整之權重調節參數。 In the aforementioned product preference combination recommendation system and method, the time period weight value of each time period interval is calculated by the following formula: W n = 1 + ( n -1) r , where W n represents the first The weight assigned to the use of the matching interval of the discount in the n time period intervals, n represents the nth time period interval after the discount is enabled, and r represents an adjustable weight adjustment parameter.
在前述之產品優惠組合推薦系統與方法中,該產品資訊處理模組係依據輸入的類別或名稱與現有產品匹配以取得使用者感興趣的該產品。 In the aforementioned product preferential combination recommendation system and method, the product information processing module is matched with an existing product according to the entered category or name to obtain the product that the user is interested in.
藉由以優惠為出發點的數據分析設計模式,透過統計過往已享優惠之使用量,結合優惠內容所反映出的價值量化數據,進行整體性的考量與評估,可快速反應市場消費脈動,有效運用於行銷推廣的電信產品優惠組合推薦;藉由產品資訊處理模組接收所輸入的產品類型或名稱資訊擷取出產品資訊集合;透過產品優惠映射模組利用產品及優惠之間的關聯規則,找出輸入產品及其對應優惠集合;透過優惠使用統計模組依據各個優惠過往已取用之歷史資訊進行分析,搭配時間週期做為資料統計及權重分配的條件,取得各時間週期內經權重分配後的優惠使用量數據; 透過優惠價值量化模組將各個優惠所贈送的內容,如實體商品、加值服務或是費用折扣等異質性項目轉換成如紅利點數、市售金額等可量化比較的價值數據;以及透過優惠組合推薦模組依據產品所關聯的優惠資料結合優惠使用統計及優惠價值數據進行評量,動態實現產品建議與優惠組合推薦之作業;當消費者所提供的感興趣產品類型或是名稱,由產品資訊處理模組產生候選產品集合,並輸入產品優惠映射模組取得個別產品對應之關聯優惠集合,再將關聯優惠集合分別傳入優惠使用統計模組及優惠價值量化模組,取得優惠集合中個別優惠的使用量統計數據及可量化比較的價值數據,之後再交由優惠組合推薦模組依照所傳入的優惠使用量及價值數據進行優惠組合推薦運算,即時且動態產出所建議的產品及其推薦優惠的組合。 Based on the data analysis and design mode based on the discount, the statistics can be used to comprehensively consider and evaluate the total consumption of discounts in the past, combined with the quantitative data of the value reflected in the discount content, to quickly reflect the market consumption pulse, and effectively use it. Recommended product portfolios for telecommunications products promoted in marketing; Product information processing module receives input product type or name information to extract product information collection; Product product mapping module uses product rule and association rules to find out Enter the product and its corresponding set of discounts; use the discount usage statistics module to analyze the historical information of each discount in the past, and use the time period as the data statistics and weight distribution conditions to obtain the weighted discounts in each time period Usage data Use the discount value quantification module to convert the contents of each discount, such as physical goods, value-added services or fee discounts, into heterogeneous items such as bonus points, market amounts, and other quantifiable and comparable value data; and The combination recommendation module evaluates according to the preferential data associated with the product in combination with preferential usage statistics and preferential value data to dynamically implement the product recommendation and preferential combination recommendation operation. When the type or name of the product of interest provided by the consumer, the product The information processing module generates a set of candidate products, and enters the product preference mapping module to obtain the associated preferential set corresponding to the individual product, and then transmits the associated preferential set to the preferential usage statistics module and the preferential value quantification module, respectively, to obtain individual items in the preferential set. Preferential usage statistics and quantifiable and comparable value data are then submitted to the Preferential Portfolio Recommendation Module to perform preferential combination recommendation calculations based on the incoming concessional usage and value data, real-time and dynamic output of the recommended products and Its recommended offer combination.
100‧‧‧產品資訊處理模組 100‧‧‧Product Information Processing Module
200‧‧‧產品優惠映射模組 200‧‧‧Product Offer Mapping Module
300‧‧‧優惠使用統計模組 300‧‧‧Offer usage statistics module
400‧‧‧優惠價值量化模組 400‧‧‧ Preferential Value Quantification Module
500‧‧‧優惠組合推薦模組 500‧‧‧Offer Combination Recommended Module
S201-S206‧‧‧步驟 S201-S206‧‧‧ steps
請參閱有關本發明之詳細說明及其附圖,將可進一步瞭解本發明之技術內容及其目的功效,有關附圖為:第1圖為本發明之產品優惠組合推薦系統之示意架構圖;以及第2圖為本發明之產品優惠組合推薦方法之示意流程圖。 Please refer to the detailed description of the present invention and its accompanying drawings for further understanding of the technical content of the present invention and its purpose and effectiveness. The related drawings are: FIG. 1 is a schematic architecture diagram of the product preferential combination recommendation system of the present invention; and FIG. 2 is a schematic flowchart of a product preference combination recommendation method of the present invention.
以下在實施方式中將詳實敘述本發明之重要特徵與優點,其內容足以使任何熟知相關技藝者瞭解本發明之技術內容並據以實施,且可輕易理解本發明之目的與優點。 The important features and advantages of the present invention will be described in detail in the following embodiments. The content is sufficient for any person skilled in the art to understand and implement the technical content of the present invention, and can easily understand the purpose and advantages of the present invention.
請參閱第1圖,係本發明之產品優惠組合推薦系統之示意架構圖,由圖中可知,本發明為一種針對優惠內容進行數據分析之電信產品優惠組合推薦系統與方法,其包括產品資訊處理模組100、產品優惠映射模組200、優惠使用統計模組300、優惠價值量化模組400及優惠組合推薦模組500,其中:產品資訊處理模組100可將輸入的產品類型或名稱與現有電信產品進行匹配,找出使用者可能感興趣的候選產品集合,再將符合條件的候選產品集合之資訊傳送至產品優惠映射模組200處理,藉以取得產品與優惠之間的關聯性。 Please refer to FIG. 1, which is a schematic architecture diagram of the product preferential combination recommendation system of the present invention. As can be seen from the figure, the present invention is a telecommunication product preferential combination recommendation system and method for analyzing data on preferential content, which includes product information processing. Module 100, product offer mapping module 200, offer usage statistics module 300, offer value quantification module 400, and offer combination recommendation module 500, of which: the product information processing module 100 can input the type or name of the product with the existing The telecommunication products are matched to find the candidate product set that the user may be interested in, and then the information of the candidate product set that meets the conditions is transmitted to the product preference mapping module 200 for processing, so as to obtain the correlation between the product and the offer.
產品優惠映射模組200透過對電信產品資訊集內每一產品辨識碼為鍵值各別找出與此產品有設定關聯且於期效內之可用優惠資訊,而產品與優惠之間可為多對多映射關係,故也可以優惠代碼為鍵值找出與此優惠有設定關聯之產品資訊。 The product offer mapping module 200 finds the available offer information associated with this product and within the period of validity by identifying each product identification code in the telecommunication product information set as a key value, and there can be more between the product and the offer. For many-to-many mapping relationship, you can also use the promotion code as a key value to find out the product information associated with this promotion.
優惠使用統計模組300從產品優惠映射模組200傳入之優惠代碼,找出該優惠代碼之使用紀錄,並結合時間週期作為資料取用區間範圍及權重分配條件進行運算,時間週期可為常用之每日、每週、每月、每年等時間週期或依需求自訂,如時間週期為每月,則統計該優惠啟用至今之每月優惠使用量,而權重分配則是依下列之公式1,對每月優惠使用量進行權重調整,如下所示:W n =1+(n-1)r (公式1), 其中,n表示優惠啟用後第n個月,r為一可調整之權重調節參數,Wn表示第n個月時優惠使用量所分配到的權重,藉由計算得到每月使用量與加權值之乘積值,再算出該優惠之每月加權平均使用量(加權平均數),作為優惠使用量資訊。 The discount usage statistics module 300 uses the discount code passed in from the product discount mapping module 200 to find out the usage history of the discount code, and combines the time period as the range of data access and weight distribution conditions for calculation. The time period can be commonly used Time periods such as daily, weekly, monthly, yearly, etc., or customized according to demand. If the time period is monthly, the monthly usage of the discount so far is counted, and the weight distribution is based on the following formula 1 , Adjust the weight of the monthly discount usage as follows: W n = 1 + ( n -1) r (Formula 1), where n is the nth month after the discount is activated, and r is an adjustable weight Adjust the parameter, W n represents the weight assigned to the preferential usage amount in the nth month. The product of the monthly usage amount and the weighted value is calculated, and then the monthly weighted average usage amount (weighted average number) of the discount is calculated. ) As discount usage information.
優惠價值量化模組400取得從產品優惠映射模組200傳入之優惠代碼,找出對應此優惠的內容,內容可為費用的折扣、實體商品或電信產品的贈送等項目組合而成,經由優惠內容與價值對應表加以轉換成如紅利點數、市售金額等可量化的優惠價值資訊。 The preferential value quantization module 400 obtains the discount code passed in from the product discount mapping module 200, and finds out the content corresponding to this discount. The content can be a combination of items such as discounts on fees, gifts of physical products or telecommunications products. The content and value correspondence table is converted into quantifiable preferential value information such as bonus points and marketed amounts.
優惠組合推薦模組500取得由優惠使用統計模組300傳來的優惠統計量資訊及由優惠價值量化模組400傳來的優惠價值資訊,再依據產品優惠映射模組200所得到的候選產品集合進行產品數量判斷,如產品數量僅有一項,則針對此候選產品所含之優惠集合,逐一計算各項優惠使用量(加權平均使用量)與優惠價值之乘積值,並依乘積值由高至低排序,列出產品與推薦之優惠組合;如產品數量多於一項,則對每個候選產品其關聯之優惠集合內所對應的優惠使用量(加權平均使用量)與優惠價值取乘積平均值,取得具最大乘積平均值的產品做為建議產品後,再依產品數量僅有一項的處理程序找出此產品與推薦之優惠組合。 The preferential combination recommendation module 500 obtains the preferential statistics information transmitted from the preferential usage statistics module 300 and the preferential value information transmitted from the preferential value quantization module 400, and then the candidate product set obtained according to the product preferential mapping module 200 Judging the number of products, if there is only one product, the product of each preferential use (weighted average use) and the value of the offer is calculated one by one for the set of offers included in this candidate product, and the product value is increased from as high as Low ranking, listing product and recommended offer combinations; if the number of products is more than one, for each candidate product, the corresponding preferential usage (weighted average usage) and preferential value in the associated preferential set are multiplied by the product average Value, get the product with the largest product average as the recommended product, and then find out the combination of this product and the recommended discount according to the processing procedure with only one item of product quantity.
為能更清楚瞭解本發明之內容、目的、特徵及功能,僅以下列案例詳述關於本發明內容及實施方式,用以示範與解釋本發明之原理。 In order to understand the content, purpose, features, and functions of the present invention more clearly, only the following examples are used to describe the content and implementation of the present invention in detail to demonstrate and explain the principle of the present invention.
初始資料(僅列舉與本發明相關的部分資料):
輸入P1產品至產品資訊處理模組100,因P1屬產品,故直接帶出產品P1資訊,再由產品優惠映射模組200至表1取得關聯優惠集{D1,D2,D3},並以優惠代碼作為優惠使用統計模組300及優惠價值量化模組400的輸入鍵值,優惠使用統計模組300根據表2進行每月優惠使用量加權平均計算得到表3,之後優惠組合推薦模組500再依照表3及優惠價值量化模組400所得結果表4,分別取得各關聯優惠統計資料及優惠價值資料進行綜合評估,最後 再依優惠使用量(加權平均使用量)與優惠價值的乘積值由高至低進行排序,如下表5,得出推薦之優惠組合為P1產品與D3優惠之組合。 Enter the P1 product into the product information processing module 100. Because P1 is a product, the product P1 information is directly taken out, and then the product discount mapping module 200 to Table 1 is used to obtain the associated discount set {D1, D2, D3}, and use the discount The code is used as the input key value of the preferential usage statistics module 300 and the preferential value quantization module 400. The preferential usage statistics module 300 performs the monthly preferential usage weighted average calculation according to Table 2 to obtain Table 3, and then the preferential combination recommendation module 500 re- According to Table 3 and the result table 4 of the preferential value quantization module 400, obtain the relevant preferential statistics and preferential value data for comprehensive evaluation, and finally Then sort by the product value of the preferential usage (weighted average usage) and the preferential value from high to low, as shown in Table 5 below. The recommended preferential combination is the combination of P1 products and D3 discounts.
輸入PL1類產品透過產品資訊處理模組100至表1取得所歸屬之候選產品集合{P1,P2,P3},再由產品優惠映射模組200至表1取得各產品的關聯優惠集合{D1,D2,D3}、{D2,D3}、{D1,D2},並以各產品的關聯優惠集合內的優惠代碼作為優惠使用統計模組300及優惠價值量化模組400的輸入鍵值,優惠使用統計模組300根據表2進行每月優惠使用量加權平均計算得到表3,而優惠價值量化模組400所得結果為表4,之後優惠組合推薦模組500再依照表3及表4分別取得優惠統計資料及優惠價值資料,最後由優惠組合推薦模組500進行評估,針對各項產品的關聯優惠集合內的各項優惠逐一運算,取得優惠使用量(加權平均使用量)與優惠價值的乘積,再將對此優惠集 合內各優惠的乘積進行加總後除以優惠數量取得乘積平均值,並以乘積平均值作為評比標的,取出具有最大乘積平均值之候選產品P2做為建議產品,如下之表6,接著同案例一的處理步驟,針對建議產品P2之優惠集合{D2,D3}內各優惠進行推薦評估,如下之表7,可得推薦的產品優惠組合為申辦PL1類產品中的P2產品與D3優惠組合。 Enter PL1 products to obtain the candidate product set {P1, P2, P3} that belongs to them through the product information processing module 100 to Table 1, and then obtain the associated discount set {D1, D2, D3}, {D2, D3}, {D1, D2}, and use the discount code in the associated discount set of each product as the input key value of the discount usage statistics module 300 and the discount value quantization module 400 for preferential use The statistical module 300 performs the weighted average calculation of the monthly preferential usage according to Table 2 to obtain Table 3, and the result obtained by the preferential value quantization module 400 is Table 4, and the discount combination recommendation module 500 then obtains the discounts in accordance with Tables 3 and 4 respectively. The statistical data and preferential value data are finally evaluated by the preferential combination recommendation module 500, and each discount in the associated discount set of each product is calculated one by one to obtain the product of preferential usage (weighted average usage) and preferential value. Will set this offer again The product of each discount in the portfolio is added up and divided by the number of discounts to obtain the product average, and the product average is used as the evaluation target. The candidate product P2 with the largest product average is taken as the recommended product, as shown in Table 6 below. The processing steps in case one are recommended evaluation for each offer in the offer set of recommended product P2 {D2, D3}, as shown in Table 7 below. The recommended product combination is the combination of P2 products and D3 offers in the PL1 category. .
請參考第2圖,為本發明之電信產品優惠組合推薦方法之示意流程圖,其步驟如下: Please refer to FIG. 2 for a schematic flowchart of a method for recommending a preferential combination of telecommunication products according to the present invention. The steps are as follows:
在步驟S201中,透過產品資訊處理模組100讀取輸 入之產品類型或名稱資訊進行解析,找出有效可供裝的候選產品資料集合。 In step S201, the output is read through the product information processing module 100. Enter the product type or name information to analyze to find a valid candidate product data set for installation.
在步驟S202中,所取得的候選產品集合之資訊會傳送到產品優惠映射模組200,針對候選產品集合內各項產品的識別碼作為鍵值進行產品優惠關聯映射,找出各項候選產品所屬的可用優惠集合。 In step S202, the obtained information of the candidate product set is transmitted to the product preference mapping module 200, and the product preference association mapping is performed on the identification codes of each product in the candidate product set as key values to find out which candidate products belong to Of available offers.
在步驟S203中,優惠使用統計模組300及優惠價值量化模組400會根據所傳入之優惠集合內各項優惠代碼作為鍵值,分別取得該優惠時間週期內的加權平均使用量與該優惠內容的價值量化數值。 In step S203, the discount usage statistics module 300 and the discount value quantization module 400 will obtain the weighted average usage amount and the discount in the discount time period according to the various discount codes in the incoming discount set as key values. Quantitative value of content.
在步驟S204中,在進行產品優惠組合推薦之前,優惠組合推薦模組500先依照所傳入的產品資料集合內的產品數量是否僅有一項產品存在進行判斷。 In step S204, before performing a product discount combination recommendation, the discount combination recommendation module 500 first determines whether there is only one product in the quantity of products in the incoming product data set.
在步驟S205中,若候選產品集合內含有多項產品,則會將候選產品集合內的各項產品逐一取出,以產品識別碼作為鍵值取得對應的關聯優惠集合,再將關聯優惠集合內的各項優惠逐一取出,以優惠代碼作為鍵值,找出對應的優惠使用量(加權平均使用量)與優惠價值並計算各項優惠所含使用量與優惠價值之乘積後,進一步算出此優惠集合內的乘積平均值,以此作為排序依據由高至低排序,取得具有最大乘積平均值的候選產品。 In step S205, if there are multiple products in the candidate product set, each product in the candidate product set will be taken out one by one, and the corresponding associated benefit set is obtained using the product identification code as a key value, and then each of the associated benefit set will be obtained. Each offer is taken out one by one, using the promotion code as a key value, to find the corresponding preferential use (weighted average use) and the value of the offer, and calculate the product of the amount of use and the value of the offer, and then further calculate the The average value of the product is used as the sorting basis from high to low to obtain the candidate product with the largest product average.
在步驟S206中,此時候選產品集合內將只保留此項候選產品進行優惠組合推薦,若候選產品集合內僅有一項產品時,則會將該產品所屬的關聯優惠集合內各項優惠逐 一取出,以優惠代碼作為鍵值,計算優惠使用量(加權平均使用量)與優惠價值的乘積值,再依照乘積值由高至低進行排序,取得推薦組合的優先次序,藉此獲得推薦的產品及優惠組合。 In step S206, at this time, only this candidate product will be retained in the candidate product set for recommendation combination recommendation. If there is only one product in the candidate product set, each benefit in the associated benefit set to which the product belongs will be listed one by one. Once taken out, use the discount code as the key value to calculate the product value of the discounted usage amount (weighted average usage amount) and the discounted value, and then sort according to the product value from high to low to obtain the priority of the recommended combination, thereby obtaining the recommended Products and offers.
本發明所提供之電信產品優惠推薦系統與方法與其他現行做法相比較之下,已具備優點如下:本發明係從已享用之優惠數據進行資料探勘為基礎,透過輸入產品類型或名稱,找出對應優惠資料集合,並以優惠使用量及優惠價值兩項數據做為評量基準,從而進一步推薦產品及優惠組合,相較一般由產品銷售量為基礎所作出之熱門產品推薦排行而言,能更反映消費者對產品及搭配優惠的消費傾向及潛在需求。 Compared with other current practices, the telecommunication product preferential recommendation system and method provided by the present invention have the following advantages: The present invention is based on data exploration from the enjoyed preferential data, and is found by entering the product type or name Corresponds to the collection of preferential data, and uses the two data of preferential usage and preferential value as the evaluation basis to further recommend products and preferential combinations. Compared with the popular product recommendation ranking based on product sales, it can It also reflects consumers ’consumption tendency and potential demand for products and matching offers.
本發明在產品及優惠之間導入優惠使用量及優惠價值兩項數值做為產品優惠組合推薦之排序衡量指標,相較習用僅由折扣金額做為優惠推薦的排序依據有所差異,可更彈性的應用在電信產品的行銷推廣上。 The present invention introduces two values of preferential usage amount and preferential value between products and offers as the ranking measurement index of product preferential combination recommendation, which is more flexible than the conventional use of only the discount amount as the basis for preferential recommendation. Application in the marketing of telecommunications products.
本發明在優惠使用量計算方面,提出時間週期權重調整技術,可隨時間演進對優惠使用量進行動態調整,有別於以往對於優惠使用量數據運用上的統計方式,能有效反映出市場所關注的趨勢潮流,更切合消費者對產品及優惠的推薦期望。因此,本發明不僅具備創新技術之巧思,並具備習用之傳統方法所不及之上述多項功效。 In the calculation of preferential usage, the present invention proposes a time period weight adjustment technology, which can dynamically adjust the preferential usage over time, which is different from the previous statistical method of using preferential usage data and can effectively reflect the market's concern. The trend is more in line with consumers' recommendation expectations for products and offers. Therefore, the present invention not only possesses the ingenuity of the innovative technology, but also has the above-mentioned multiple effects which cannot be achieved by the conventional traditional methods.
上述實施例係用以例示性說明本發明之原理及其功效,而非用於限制本發明。任何熟習此項技藝之人士均可 在不違背本發明之精神及範疇下,對上述實施例進行修改。因此本發明之權利保護範圍,應如後述之申請專利範圍所列。 The above embodiments are used to exemplify the principle of the present invention and its effects, but not to limit the present invention. Anyone who is familiar with this skill The above embodiments are modified without departing from the spirit and scope of the present invention. Therefore, the scope of protection of the rights of the present invention should be listed in the scope of patent application described later.
Claims (8)
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
TW107103633A TWI652639B (en) | 2018-02-01 | 2018-02-01 | Recommended system and method of product promotion combination |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
TW107103633A TWI652639B (en) | 2018-02-01 | 2018-02-01 | Recommended system and method of product promotion combination |
Publications (2)
Publication Number | Publication Date |
---|---|
TWI652639B TWI652639B (en) | 2019-03-01 |
TW201935363A true TW201935363A (en) | 2019-09-01 |
Family
ID=66590578
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
TW107103633A TWI652639B (en) | 2018-02-01 | 2018-02-01 | Recommended system and method of product promotion combination |
Country Status (1)
Country | Link |
---|---|
TW (1) | TWI652639B (en) |
Families Citing this family (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111738786A (en) * | 2019-04-24 | 2020-10-02 | 北京京东尚科信息技术有限公司 | Method, system, apparatus and readable storage medium for building a combination of commodities |
CN110555752A (en) * | 2019-07-31 | 2019-12-10 | 北京三快在线科技有限公司 | Personalized recommendation method and device, electronic equipment and readable storage medium |
CN111339408B (en) * | 2020-02-19 | 2023-08-08 | 北京百度网讯科技有限公司 | Information recommendation method and device |
CN112102032A (en) * | 2020-09-01 | 2020-12-18 | 北京思特奇信息技术股份有限公司 | Service processing method and related system |
JP7170785B1 (en) * | 2021-05-13 | 2022-11-14 | 楽天グループ株式会社 | Information processing system, information processing method and program |
TWI828128B (en) * | 2022-04-28 | 2024-01-01 | 中華電信股份有限公司 | A rental matching system and method for subscription service and computer-readable medium thereof |
Family Cites Families (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
TW201409374A (en) | 2012-08-20 | 2014-03-01 | Chunghwa Telecom Co Ltd | Dynamic combination method for determining pricing, selling price and commission of telecommunication terminal products |
TW201516925A (en) | 2013-10-30 | 2015-05-01 | Chunghwa Telecom Co Ltd | Assembly and charge system of processing composite discount and the method thereof |
TWI626609B (en) | 2014-01-16 | 2018-06-11 | Chunghwa Telecom Co Ltd | Production method and system for personal product and vaccine purchase combination |
-
2018
- 2018-02-01 TW TW107103633A patent/TWI652639B/en not_active IP Right Cessation
Also Published As
Publication number | Publication date |
---|---|
TWI652639B (en) | 2019-03-01 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Rao et al. | Return time leniency in online retail: a signaling theory perspective on buying outcomes | |
Huang et al. | An empirical study of the cross-channel effects between web and mobile shopping channels | |
TW201935363A (en) | Recommended system and method of product promotion combination | |
JP5394915B2 (en) | Distributed architecture for online advertising | |
Kuusik | Affecting customer loyalty: Do different factors have various influences in different loyalty levels? | |
Fang et al. | Direct and indirect effects of buyers and sellers on search advertising revenues in business-to-business electronic platforms | |
US20080103887A1 (en) | Selecting advertisements based on consumer transactions | |
US20080103897A1 (en) | Normalizing and tracking user attributes for transactions in an advertising exchange | |
Omotayo et al. | Determinants of continuance intention to use online shops in Nigeria | |
US20080103896A1 (en) | Specifying, normalizing and tracking display properties for transactions in an advertising exchange | |
US20110099060A1 (en) | System and method for dynamic price setting and facilitation of commercial transactions | |
Wilis et al. | The Effect of Digital Marketing, Influencer Marketing and Online Customer Review on Purchase Decision: A Case Study of Cake Shop" Lu’miere" | |
Leeflang et al. | Identifying the direct mail-prone consumer | |
Hertina et al. | Analysis of buying decision levels based on brand image, price, and digital marketing | |
US20240112210A1 (en) | Self-learning valuation | |
WO2002079901A2 (en) | Customer preference system | |
Duch-Brown | Platforms to business relations in online platform ecosystems | |
Dharmawan et al. | Customer-based brand equity in digital age: A conceptual approach | |
Pathan et al. | How digital marketing influences purchase intention? A case of fast-food industry | |
Sanny et al. | The impact of instagram e-marketing in SME Fashion Industry on customer satisfaction | |
Carr et al. | Measuring customer equity and managing the customer asset: An explanatory study in the Australian hotels | |
Garfinkel et al. | Empirical analysis of the business value of recommender systems | |
Mofokeng | Website Information Security and Privacy Concerns in 4IR: The Moderating Role of Trust in B2C e-Commerce. | |
Thi et al. | Mediating effect of social commerce continuance use intention on the relationship between perceived values and brand loyalty | |
Jiang | Understanding the efficiency of keywords in search engine marketing-a comparison study using data envelopment analysis |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
MM4A | Annulment or lapse of patent due to non-payment of fees |