TWI769385B - Method and system for screening potential purchasers of financial products - Google Patents
Method and system for screening potential purchasers of financial products Download PDFInfo
- Publication number
- TWI769385B TWI769385B TW108123422A TW108123422A TWI769385B TW I769385 B TWI769385 B TW I769385B TW 108123422 A TW108123422 A TW 108123422A TW 108123422 A TW108123422 A TW 108123422A TW I769385 B TWI769385 B TW I769385B
- Authority
- TW
- Taiwan
- Prior art keywords
- customer
- recommended
- previous period
- ratio
- client
- Prior art date
Links
Images
Landscapes
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
一種金融商品之潛在購買客群篩選方法,藉由一電腦裝置來實施,該電腦裝置儲存有多筆相關於多個客戶的信用能力資料、多筆相關於該等客戶於一先前期間內之財務狀況的總歷史資產資料,以及多個不利條件。該方法包含以下步驟:該電腦裝置根據客戶所對應的信用能力資料及總歷史資產資料,判定客戶是否為高消費力客戶;該電腦裝置根據高消費力客戶所對應的總歷史資產資料及該等不利條件,判定高消費力客戶是否為欲推薦客戶;該電腦裝置根據欲推薦客戶的總歷史資產資料,利用客戶分類模型,將欲推薦客戶歸類為推薦客戶類別或不推薦客戶類別。 A method for screening potential purchasers of financial products, implemented by a computer device, the computer device storing a plurality of pieces of credit ability data related to a plurality of customers, and a plurality of pieces of financial data related to the customers in a previous period Total historical asset information for conditions, as well as multiple adverse conditions. The method includes the following steps: the computer device determines whether the customer is a high spending power customer according to the credit ability data and the total historical asset data corresponding to the customer; the computer device determines whether the customer is a high spending power customer according to the total historical asset data corresponding to the high spending power customer and the Unfavorable conditions, determine whether a customer with high spending power is a customer to be recommended; the computer device uses a customer classification model to classify the customer to be recommended as a recommended customer category or a non-recommended customer category according to the total historical asset data of the recommended customer.
Description
本發明是有關於一種客戶分析系統,特別是指一種相關於金融領域的客戶篩選系統。The present invention relates to a customer analysis system, in particular to a customer screening system related to the financial field.
目前於金融領域的銀行業或是保險業所發行的金融產品層出不窮,因此越來越多相關的研發人員利用各種AI人工智慧技術對所蒐集的客戶資料進行分析篩選,藉此獲得客戶是否具有足夠消費能力及購買金融商品的可能性等種種資訊之結果,好以根據分析結果推銷各種金融產品至合適的客戶。At present, there are many financial products issued by the banking or insurance industry in the financial field. Therefore, more and more relevant R&D personnel use various AI artificial intelligence technologies to analyze and filter the collected customer data, so as to obtain whether the customer has sufficient The results of various information such as spending power and the possibility of purchasing financial products, so as to market various financial products to appropriate customers based on the analysis results.
然而,實務上,單單地將客戶資料利用AI人工智慧進行分析後所獲得的結果,即使換用各種現有的AI人工智慧技術,其準確率卻仍無法達到所預期的效果,而準確率不佳的分析結果往往造成銀行業或是保險業無法找到合適的客戶,以致於所發行的金融產品推廣不易且效果不彰。However, in practice, the results obtained after simply analyzing customer data with AI artificial intelligence, even if various existing AI artificial intelligence technologies are used, the accuracy rate still cannot achieve the expected results, and the accuracy rate is not good. The results of the analysis often result in the banking industry or the insurance industry being unable to find suitable customers, so that the promotion of the issued financial products is difficult and ineffective.
因此,故如何提出一種能搭配AI人工智慧並使所獲得的分析結果更為準確的方法及系統,即為本創作所欲解決之首要課題。Therefore, how to propose a method and system that can match AI artificial intelligence and make the obtained analysis results more accurate is the primary issue that this creation intends to solve.
因此,本發明的目的,即在提供一種將輸入資料進行前處理,並搭配AI人工智慧使所獲得的分析結果更為準確的金融商品之潛在購買客群篩選方法。Therefore, the purpose of the present invention is to provide a method for screening potential purchasers of financial products that pre-processes the input data and uses AI artificial intelligence to obtain more accurate analysis results.
於是,本發明金融商品之潛在購買客群篩選方法,藉由一電腦裝置來實施,該電腦裝置儲存有多筆相關於多個客戶之信用的信用能力資料、多筆相關於該等客戶於一先前時間點至一當前時間點所界定之一先前期間內之財務狀況的總歷史資產資料,以及多個相關於各種產業及客戶經濟狀況的不利條件,並包含一步驟(A)、一步驟(B)、一步驟(C)。Therefore, the screening method for potential purchasers of financial products of the present invention is implemented by a computer device, the computer device stores a plurality of pieces of credit ability data related to the credit of a plurality of customers, and a plurality of pieces of credit ability data related to the customers in a Aggregate historical asset information on the financial position of a prior period defined from a previous point in time to a current point in time, and a number of adverse conditions related to various industry and customer economic conditions, including a step (A), a step ( B), a step (C).
步驟(A)是,對於每一客戶,藉由該電腦裝置,根據該客戶所對應的該信用能力資料及該總歷史資產資料,判定該客戶是否為一高消費力客戶。In step (A), for each customer, determine whether the customer is a high spending power customer according to the credit capability data and the total historical asset data corresponding to the customer through the computer device.
步驟(B)是,對於每一高消費力客戶,藉由該電腦裝置,根據該高消費力客戶所對應的該總歷史資產資料及該等不利條件,判定該高消費力客戶是否為一欲推薦客戶。Step (B) is, for each high spending power customer, by the computer device, according to the total historical asset data and the unfavorable conditions corresponding to the high spending power customer, determine whether the high spending power customer is a desire Recommend clients.
步驟(C)是,對於每一欲推薦客戶,藉由該電腦裝置,根據該欲推薦客戶所對應的該總歷史資產資料,利用一用於將客戶分類為推薦客戶或不推薦客戶的客戶分類模型,將該欲推薦客戶歸類為一推薦客戶類別及一不推薦客戶類別之其中一者。Step (C) is, for each customer to be recommended, using the computer device to use a customer classification for classifying the customer as a recommended customer or a non-recommended customer according to the total historical asset data corresponding to the customer to be recommended The model classifies the customer to be recommended as one of a recommended customer category and a non-recommended customer category.
本發明之另一目的,即在提供一種將輸入資料進行前處理,並搭配AI人工智慧使所獲得的分析結果更為準確的金融商品之潛在購買客群篩選系統。Another object of the present invention is to provide a potential buyer group screening system for financial products that pre-processes the input data and uses AI artificial intelligence to obtain more accurate analysis results.
於是,本發明金融商品之潛在購買客群篩選系統,包含一儲存模組,以及一電連接該儲存模組的處理模組。Therefore, the system for screening potential purchasers of financial products of the present invention includes a storage module and a processing module electrically connected to the storage module.
該儲存模組儲存有多筆相關於多個客戶之信用的信用能力資料、多筆相關於該等客戶於一先前時間點至一當前時間點所界定之一先前期間內之財務狀況的總歷史資產資料,以及多個相關於各種產業及客戶經濟狀況的不利條件。The storage module stores a plurality of pieces of credit capability data related to the credit of a plurality of customers, a plurality of pieces of total history related to the financial conditions of the customers from a previous point in time to a previous period defined by a current point in time Asset information, as well as a number of adverse conditions related to various industries and customer economic conditions.
其中,對於每一客戶,該處理模組根據該客戶所對應的該信用能力資料及該總歷史資產資料,判定該客戶是否為一高消費力客戶,對於每一高消費力客戶,該處理模組根據該高消費力客戶所對應的該總歷史資產資料及該等不利條件,判定該高消費力客戶是否為一欲推薦客戶,對於每一欲推薦客戶,該處理模組根據該欲推薦客戶所對應的該總歷史資產資料,利用一用於將客戶分類為推薦客戶或不推薦客戶的客戶分類模型,將該欲推薦客戶歸類為一推薦客戶類別及一不推薦客戶類別之其中一者。Wherein, for each customer, the processing module determines whether the customer is a high spending power customer according to the credit ability data and the total historical asset data corresponding to the customer, and for each high spending power customer, the processing module The group determines whether the high spending power customer is a customer to be recommended according to the total historical asset data corresponding to the high spending power customer and the unfavorable conditions. For each customer to be recommended, the processing module determines whether the customer is to be recommended The corresponding total historical asset data, using a customer classification model for classifying customers as recommended customers or non-recommended customers, classify the customer to be recommended as one of a recommended customer category and a non-recommended customer category .
本發明之功效在於:在利用該客戶分類模型(AI人工智慧)進行分類之前,先藉由該電腦裝置執行一能搭配該客戶分類模型的資料前處理,也就是說,首先,根據該客戶所對應的該信用能力資料及該總歷史資產資料,進行高消費力客戶的篩選,接著,根據該高消費力客戶所對應的該總歷史資產資料及該等不利條件,進行欲推薦客戶篩選,當資料前處理完成後,才將該欲推薦客戶所對應的總歷史資產資料透過該客戶分類模型(AI人工智慧)進行處理,以達到相較於傳統僅直接用該客戶分類模型(AI人工智慧)進行分類更加準確地的分類出推薦/不推薦客戶類別的結果,以更有效地推廣金融產品。The effect of the present invention is: before using the customer classification model (AI artificial intelligence) for classification, the computer device performs a data preprocessing that can match the customer classification model, that is, first, according to the customer Corresponding to the credit ability data and the total historical asset data, perform screening for customers with high spending power, and then, according to the total historical asset data and the disadvantageous conditions corresponding to the high spending power customers, screen the customers to be recommended. After the data pre-processing is completed, the total historical asset data corresponding to the customer to be recommended is processed through the customer classification model (AI artificial intelligence), so as to achieve the goal of directly using the customer classification model (AI artificial intelligence) compared to the traditional method. Perform classification to more accurately classify the results of recommended/not recommended customer categories, so as to promote financial products more effectively.
在本發明被詳細描述之前,應當注意在以下的說明內容中,類似的元件是以相同的編號來表示。Before the present invention is described in detail, it should be noted that in the following description, similar elements are designated by the same reference numerals.
參閱圖1,本發明金融商品之潛在購買客群篩選系統之一實施例包含一電腦裝置1。該電腦裝置1包含一儲存模組11、一顯示模組12,以及一電連接該儲存模組11及該顯示模組12的處理模組13。Referring to FIG. 1 , an embodiment of the system for screening potential purchasers of financial products of the present invention includes a
該儲存模組11儲存有多筆相關於多個客戶之信用的信用能力資料、多筆相關於該等客戶於一先前時間點至一當前時間點所界定之一先前期間內之財務狀況的總歷史資產資料、多個相關於各種產業及客戶經濟狀況的不利條件,以及一用於將客戶分類為推薦客戶或不推薦客戶的客戶分類模型。其中,每一客戶屬於一企業戶及一個人戶之其中一者。The
每筆屬於該企業戶的客戶之總歷史資產資料包含該客戶在該先前期間之多個不同時間點所對應的多筆毛利率、該客戶在該先前期間之多個不同時間點所對應的多筆營業收入、該客戶在該先前期間之多個不同時間點所對應的多筆收帳天期、該客戶對應之產業在該先前期間之多個不同時間點所對應的多筆景氣衰退比率、該客戶在該先前期間之多個不同時間點所對應的多筆營收與獲利比率、該客戶在該先前期間之多個不同時間點所對應的多筆營收資訊、該客戶在該先前期間之多個不同時間點所對應的多筆稅前損益、該客戶在該先前期間之多個不同時間點所對應的多筆收帳天期、該客戶在該先前期間之多個不同時間點所對應的多筆產業景氣指標(指示出當下時間點景氣為衰退、進步或持平),以及該客戶在該先前期間之多個不同時間點所對應的多筆現金流量比。The total historical asset data of each customer belonging to the enterprise account includes multiple gross profit margins for the customer at different time points in the previous period, multiple gross profit margins for the customer at different time points in the previous period operating income, multiple collection days corresponding to the customer at multiple different time points in the previous period, multiple business recession ratios corresponding to the customer’s industry at multiple different time points in the previous period, Multiple revenue-to-profit ratios for the customer at multiple different time points in the previous period, multiple revenue information for the customer at multiple different time points in the previous period, Multiple pre-tax profit and loss corresponding to multiple different time points in the period, multiple collection days corresponding to the customer at multiple different time points in the previous period, multiple different time points for the customer in the previous period Corresponding multiple industry prosperity indicators (indicating whether the current time point is recession, progress or flat), and multiple cash flow ratios corresponding to the customer at multiple different time points in the previous period.
每筆屬於該個人戶的客戶之總歷史資產資料包含該客戶在該先前期間之多個不同時間點所對應的多筆動產總額、該客戶在該先前期間之多個不同時間點所對應的多筆不動產總額、該客戶在該先前期間之多個不同時間點所對應的多筆個人收入、該客戶在該先前期間之多個不同時間點之所購買之高風險金融商品與所購買之全部金融產品的多筆高風險投資比率,以及該客戶在該先前期間之多個不同時間點所對應的多筆資力總額、該客戶在該先前期間之多個不同時間點所對應的多筆低風險投資資訊,以及該客戶在該先前期間之多個不同時間點所對應的多筆高風險投資資訊。其中,該等低風險投資資訊包含該客戶在該先前期間之多個不同時間點所對應的多筆儲蓄險投資資訊、該客戶在該先前期間之多個不同時間點所對應的多筆基金投資資訊、該客戶在該先前期間之多個不同時間點所對應的多筆股票投資資訊;而該等高風險投資資訊包含該客戶在該先前期間之多個不同時間點所對應的多筆目標可贖回遠期契約投資資訊(TRF,Target Redemption Forward)。The total historical asset data of each customer belonging to the individual account includes the total amount of movable properties of the customer at different time points in the previous period, the total amount of movable assets corresponding to the customer at different time points in the previous period. The total amount of real estate, the personal income of the customer at different time points in the previous period, the high-risk financial products purchased by the customer at different time points in the previous period, and all the financial products purchased The ratio of high-risk investments for the product, and the client's total capital at various points in the previous period, and the client's low-risk investments at various points in the previous period information, and information on multiple high-risk investments by the client at various points in the prior period. Wherein, the low-risk investment information includes multiple savings insurance investment information of the client at multiple different time points in the previous period, and multiple fund investments corresponding to the client at multiple different time points in the previous period Information, multiple stock investment information corresponding to the client at multiple different time points in the previous period; and such high-risk investment information includes multiple target investment information corresponding to the client at multiple different time points in the previous period. Redemption Forward Contract Investment Information (TRF, Target Redemption Forward).
該等不利條件包含多個對應於該企業戶,且分別與一預設收帳天期、一預設景氣衰退比率、一預設營收與獲利比率相關的企業戶不利條件,以及多個對應於該個人戶,且分別與一預設高風險投資比率、一預設資力比率相關的個人戶不利條件。The unfavorable conditions include a plurality of unfavorable conditions for the business account corresponding to the business account and respectively related to a predetermined collection date, a predetermined recession ratio, a predetermined revenue-to-profit ratio, and a plurality of Disadvantages of the individual account corresponding to the individual account and respectively related to a preset high-risk investment ratio and a preset capital ratio.
特別地,在本實施例中,該信用能力資料係為企業信用風險指標TCRI等級,或是為信用評級要素資料(例如:3F5C要素,3F係為經濟要素、管理要素、財務要素;5C要素係為道德品質、還款能力、資本實力、擔保和經營環境條件),但不以此為限。In particular, in this embodiment, the credit capability data is the corporate credit risk index TCRI level, or is credit rating element data (for example: 3F5C elements, 3F is economic elements, management elements, financial elements; 5C elements are moral character, repayment ability, capital strength, guarantees and business environment conditions), but not limited thereto.
特別地,在本實施例中,該客戶分類模型係利用屬於該企業戶之多個客戶的多筆企業戶訓練資料組及其對應之推薦結果,以及屬於該個人戶之多個客戶的多筆個人戶訓練資料組及其對應之推薦結果所訓練出。其中,每一企業戶所對應的企業戶訓練資料組包含在該先前期間之多個不同時間點所對應的多筆訓練營收資訊、該先前期間之多個不同時間點所對應的多筆訓練稅前損益、該先前期間之多個不同時間點所對應的多筆訓練收帳天期、該先前期間之多個不同時間點所對應的多筆訓練產業景氣指標、該先前期間之多個不同時間點所對應的多筆訓練現金流量比;而每一個人戶所對應的個人戶訓練資料組包含在該先前期間之多個不同時間點所對應的多筆訓練儲蓄險投資資訊、該先前期間之多個不同時間點所對應的多筆訓練基金投資資訊、該先前期間之多個不同時間點所對應的多筆訓練股票投資資訊,以及該先前期間之多個不同時間點所對應的多筆訓練目標可贖回遠期契約投資資訊。在本實施例中,該客戶分類模型係為一監督式模型。其中,該監督式模型係為一類神經網路模型、一邏輯迴歸模型(Logistic Regression)、一決策樹模型(Decision tree)、一支援向量機(SVM ,Support Vector Machines)或一徑向基函數核(Radial basis function),但不以此為限。In particular, in this embodiment, the customer classification model utilizes a plurality of business customer training data groups belonging to a plurality of customers of the business customer and their corresponding recommendation results, as well as a plurality of customer data groups belonging to the individual customer. Trained from individual household training data sets and their corresponding recommendation results. Wherein, the enterprise training data set corresponding to each enterprise includes multiple training revenue information corresponding to multiple different time points in the previous period, and multiple training data corresponding to multiple different time points in the previous period. Pre-tax profit and loss, multiple training collection days corresponding to multiple different time points in the previous period, multiple training industry prosperity indicators corresponding to multiple different time points in the previous period, multiple different training periods corresponding to the previous period The multiple training cash flow ratios corresponding to the time point; and the individual account training data group corresponding to each individual household includes the multiple training savings insurance investment information corresponding to multiple different time points in the previous period, the Multiple training fund investment information corresponding to multiple different time points, multiple training stock investment information corresponding to multiple different time points in the previous period, and multiple training sessions corresponding to multiple different time points in the previous period Target redeemable forward contract investment information. In this embodiment, the customer classification model is a supervised model. Wherein, the supervised model is a type of neural network model, a logistic regression model (Logistic Regression), a decision tree model (Decision tree), a support vector machine (SVM, Support Vector Machines) or a radial basis function kernel (Radial basis function), but not limited to this.
在本實施例中,該電腦裝置1之實施態樣例如為一個人電腦、一伺服器或一雲端主機,但不以此為限。In this embodiment, the implementation form of the
參閱圖2~4,以下將藉由本發明金融商品之潛在購買客群篩選系統執行一金融商品之潛在購買客群篩選方法來說明該電腦裝置1之該儲存模組11、該顯示模組12,以及該處理模組13各元件的運作細節,該金融商品之潛在購買客群篩選方法包含一步驟51、一步驟52,以及一步驟53。Referring to FIGS. 2 to 4 , the
在步驟51中,對於每一客戶,該處理模組13根據該客戶所對應的該信用能力資料及該總歷史資產資料,判定該客戶是否為一高消費力客戶;當該處理模組13判定出該客戶為該高消費力客戶時,進行流程步驟52;當該處理模組13判定出該客戶不為該高消費力客戶時,結束該金融商品之潛在購買客群篩選方法。In
如圖3所示,步驟51還進一步包含一子步驟511,以及一子步驟512。As shown in FIG. 3 ,
在子步驟511中,對於每一客戶,該處理模組13根據該客戶所對應的該總歷史資產資料,獲得一指示出該客戶在該先前期間中之一第一時間區間之財務增減的財務收益資訊。其中,該第一時間區間係為近一年之時間區間,但不以此為限。In
如圖4所示,子步驟511還進一步包含一子步驟511A、一子步驟511B、一子步驟511C,以及一子步驟511D。As shown in FIG. 4 , the
在子步驟511A中,對於每一客戶,當該客戶屬於該企業戶時,該處理模組13根據該客戶所對應的該總歷史資產資料,獲得一毛利率比率及一營業收入比率,該毛利率比率係為該客戶在該先前期間中之該第一時間區間內的所有毛利率之平均除以該客戶對應之產業中該客戶以外之所有企業於該第一時間區間之所有毛利率之平均,該營業收入比率係為該客戶在該先前期間中之該第一時間區間內的最後一個時間點所對應的營業收入除以該第一時間區間內的第一個時間點所對應的營業收入。特別地,該處理模組13係先自儲存模組11中獲得該客戶對應之產業中該客戶以外之所有企業的所有毛利率,再計算出該客戶對應之產業中該客戶以外之所有企業於該第一時間區間之所有毛利率之平均。In
在子步驟511B中,對於每一客戶,當該客戶屬於該企業戶時,該處理模組13將該毛利率比率及該營業收入比率作為該財務收益資訊。In
在子步驟511C中,對於每一客戶,當該客戶屬於該個人戶時,該處理模組13根據該客戶所對應的該總歷史資產資料,獲得一可用動產金額及一個人收入比率,該可用動產金額係為該客戶在該先前期間中之該第一時間區間內的所有動產總額之平均,該個人收入比率係為該客戶在該先前期間中之該第一時間區間內的最後一個時間點所對應的個人收入除以該第一時間區間內的第一個時間點所對應的個人收入。在其他實施例中,該可用動產金額可先轉為一對應的級距或分數,再執行後序流程步驟。In
在子步驟511D中,對於每一客戶,當該客戶屬於該個人戶時,該處理模組13將該可用動產金額及該個人收入比率作為該財務收益資訊。In
在子步驟512中,對於每一客戶,該處理模組13根據該客戶所對應的該信用能力資料及該財務收益資訊,判定該客戶是否為該高消費力客戶。當該處理模組13判定出該客戶為該高消費力客戶時,進行流程步驟52;當該處理模組13判定出該客戶不為該高消費力客戶時,結束該金融商品之潛在購買客群篩選方法。In
值得特別說明的是,在本實施例之子步驟512中,對於每一客戶,當該客戶屬於該企業戶時,該處理模組13根據該客戶所對應的該信用能力資料、該毛利率比率及該營業收益比率,判定屬於該企業戶的該客戶是否為該高消費力客戶。舉例來說,當該處理模組13判定出屬於該企業戶的該客戶信用能力資料所指示出分數或等級、該毛利率比率及該營業收益比率皆超過各自對應的預設閾值,則屬於該企業戶的客戶即為該高消費力客戶。It is worth noting that, in the
值得特別說明的是,在本實施例之子步驟512中,對於每一客戶,當該客戶屬於該個人戶時,該處理模組13根據該客戶所對應的該信用能力資料、該可用動產金額及該個人收入比率,判定屬於該個人戶的該客戶是否為該高消費力客戶。舉例來說,當該處理模組13判定出屬於該企業戶的該客戶信用能力資料所指示出分數或等級、該可用動產金額及該個人收入比率皆超過各自對應的預設閾值,則屬於該個人戶的客戶即為該高消費力客戶。It is worth noting that, in the
在步驟52中,對於每一高消費力客戶,該處理模組13根據該高消費力客戶所對應的該總歷史資產資料及該等不利條件,判定該高消費力客戶是否為一欲推薦客戶。當該處理模組13判定出該高消費力客戶為該欲推薦客戶時,進行流程步驟53;當該處理模組13判定出該高消費力客戶不為該欲推薦客戶時,結束該金融商品之潛在購買客群篩選方法。In
值得特別說明的是,在本實施例之步驟52中,對於每一高消費力客戶,當該高消費力客戶屬於該企業戶時,該處理模組13係藉由判定在該先前期間中之一第二時間區間內的所有收帳天期、所有景氣衰退比率與所有營收與獲利比率之任一者是否符合該等企業戶不利條件之其中一者,以判定出該高消費力客戶是否為該欲推薦客戶。其中,當該處理模組13判定出符合該等企業戶不利條件之其中一者時,則該高消費力客戶不為該欲推薦客戶。該第二時間區間係為近一年之時間區間,但不以此為限。而在其他實施例中,該處理模組13亦可由判定在該第二時間區間內的所有收帳天期、所有景氣衰退比率與所有營收與獲利比率是否符合該等企業戶不利條件之其中任兩者,以判定出該高消費力客戶是否為該欲推薦客戶。It is worth noting that, in
值得特別說明的是,在本實施例之步驟52中,對於每一高消費力客戶,當該高消費力客戶屬於該個人戶時,該處理模組13將該先前期間中之一第三時間區間內的該資力總額之平均除以該高消費力客戶在該先前期間中之一第四時間區間內的該資力總額之平均以獲得一資力比率。接著,對於每一高消費力客戶,當該高消費力客戶屬於該個人戶時,該處理模組13係藉由判定在該先前期間中之一第五時間區間內的所有高風險投資比率與該資力比率之任一者是否符合該等個人戶不利條件之其中一者,以判定出該高消費力客戶是否為該欲推薦客戶。其中,當該處理模組13判定出符合該等個人戶不利條件之其中一者時,則該高消費力客戶不為該欲推薦客戶。該第三時間區間係為近一年之時間區間、該第四時間區間係為近五年之時間區間,而該第五時間區間係為近一年之時間區間,但皆不以此為限。而在其他實施例中,該處理模組13亦可由判定在該第五時間區間內的所有高風險投資比率與該資力比率是否符合該等個人戶不利條件之其中任兩者,以判定出該高消費力客戶是否為該欲推薦客戶。It is worth noting that, in
在步驟53中,對於每一欲推薦客戶,該處理模組13根據該欲推薦客戶所對應的該總歷史資產資料,利用用於將客戶分類為推薦客戶或不推薦客戶的該客戶分類模型,將該欲推薦客戶歸類為一推薦客戶類別及一不推薦客戶類別之其中一者並顯示於該顯示模組12。In
值得特別說明的是,在本實施例之步驟53中,對於每一欲推薦客戶,當欲推薦客戶屬於該企業戶時,該處理模組13係根據該欲推薦客戶所對應的等營收資訊、該等稅前損益、該等收帳天期、該等產業景氣指標,以及該等現金流量比,利用該客戶分類模型,將該欲推薦客戶歸類為該推薦客戶類別或該不推薦客戶類別並顯示於該顯示模組12。It is worth noting that, in
值得特別說明的是,在本實施例之步驟53中,對於每一欲推薦客戶,當欲推薦客戶屬於該個人戶時,該處理模組13係根據該欲推薦客戶所對應的該等蓄險投資資訊、該等基金投資資訊、該等股票投資資訊,以及該等目標可贖回遠期契約投資資訊,利用該客戶分類模型,將該欲推薦客戶歸類為該推薦客戶類別或該不推薦客戶類別並顯示於該顯示模組12。It is worth noting that, in
綜上所述,本發明金融商品之潛在購買客群篩選系統,藉由該處理模組13執行兩階段的資料前處理,將屬於該企業戶或該個人戶的客戶,根據各自對應的該總歷史資產資料和該等不利條件,於第一階段中篩選出該高消費力客戶,而於第階段中篩選出該欲推薦客戶,在經過兩階段的資料前處理後,搭配該客戶分類模型(AI人工智慧)將該欲推薦客戶所對應的總歷史資產資料進行處理,相較於傳統僅直接用已訓練好的各種監督式模型(AI人工智慧)進行分類,更能夠精準地將該等客戶分類為推薦/不推薦客戶類別,以有效地避免金融產品推薦於錯誤客戶而造成呆帳,及能隨時因應客戶的產業環境和營收等狀態變化將客戶進行分類,進而獲得更加準確地分類結果,便能根據分類結果更有效地推銷金融產品至合適的客戶。因此,故確實能達成本發明的目的。To sum up, in the system for screening potential purchasers of financial products of the present invention, the
惟以上所述者,僅為本發明的實施例而已,當不能以此限定本發明實施的範圍,凡是依本發明申請專利範圍及專利說明書內容所作的簡單的等效變化與修飾,皆仍屬本發明專利涵蓋的範圍內。However, the above are only examples of the present invention, and should not limit the scope of implementation of the present invention. Any simple equivalent changes and modifications made according to the scope of the patent application of the present invention and the contents of the patent specification are still included in the scope of the present invention. within the scope of the invention patent.
1:電腦裝置
11:儲存模組
12:顯示模組
13:處理模組
51~53:步驟
511~512:子步驟
511A~511D:子步驟
1: Computer device
11: Storage Module
12: Display module
13:
本發明的其他的特徵及功效,將於參照圖式的實施方式中清楚地呈現,其中: 圖1是一方塊圖,說明本發明金融商品之潛在購買客群篩選系統的一實施例; 圖2是一流程圖,說明該實施例所執行之一金融商品之潛在購買客群篩選方法; 圖3是一流程圖,說明該金融商品之潛在購買客群篩選方法如何判定一客戶是否為一高消費力客戶的細部流程;及 圖4是一流程圖,說明金融商品之潛在購買客群篩選方法如何獲得一財務收益資訊的細部流程。Other features and effects of the present invention will be clearly presented in the embodiments with reference to the drawings, wherein: FIG. 1 is a block diagram illustrating an embodiment of the screening system for potential purchasers of financial products according to the present invention; FIG. 2 is a flow chart illustrating a method for screening potential purchasers of financial products implemented in this embodiment; Fig. 3 is a flow chart illustrating the detailed process of how the screening method for potential purchasers of the financial product determines whether a customer is a high spending power customer; and FIG. 4 is a flow chart illustrating a detailed process of how the method for screening potential purchasers of financial products obtains financial return information.
51~53:步驟 51~53: Steps
Claims (12)
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
TW108123422A TWI769385B (en) | 2019-07-03 | 2019-07-03 | Method and system for screening potential purchasers of financial products |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
TW108123422A TWI769385B (en) | 2019-07-03 | 2019-07-03 | Method and system for screening potential purchasers of financial products |
Publications (2)
Publication Number | Publication Date |
---|---|
TW202103085A TW202103085A (en) | 2021-01-16 |
TWI769385B true TWI769385B (en) | 2022-07-01 |
Family
ID=75234790
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
TW108123422A TWI769385B (en) | 2019-07-03 | 2019-07-03 | Method and system for screening potential purchasers of financial products |
Country Status (1)
Country | Link |
---|---|
TW (1) | TWI769385B (en) |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106600369A (en) * | 2016-12-09 | 2017-04-26 | 广东奡风科技股份有限公司 | Real-time recommendation system and method of financial products of banks based on Naive Bayesian classification |
CN107436866A (en) * | 2016-05-25 | 2017-12-05 | 阿里巴巴集团控股有限公司 | The recommendation method and device of finance product |
TW201913516A (en) * | 2017-08-28 | 2019-04-01 | 楊少銘 | Automatic financial commodity investment analysis and decision system and method including a database, a good investment mechanism, a friendly interface, a social unit, and a digital asset voucher module |
-
2019
- 2019-07-03 TW TW108123422A patent/TWI769385B/en active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107436866A (en) * | 2016-05-25 | 2017-12-05 | 阿里巴巴集团控股有限公司 | The recommendation method and device of finance product |
CN106600369A (en) * | 2016-12-09 | 2017-04-26 | 广东奡风科技股份有限公司 | Real-time recommendation system and method of financial products of banks based on Naive Bayesian classification |
TW201913516A (en) * | 2017-08-28 | 2019-04-01 | 楊少銘 | Automatic financial commodity investment analysis and decision system and method including a database, a good investment mechanism, a friendly interface, a social unit, and a digital asset voucher module |
Also Published As
Publication number | Publication date |
---|---|
TW202103085A (en) | 2021-01-16 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
TW530234B (en) | Methods and systems for efficiently sampling portfolios for optimal underwriting | |
TW530235B (en) | Valuation prediction models in situations with missing inputs | |
CN109711955B (en) | Poor evaluation early warning method and system based on current order and blacklist base establishment method | |
US20200043022A1 (en) | Artificial intelligence system and method for generating a hierarchical data structure | |
CN111861698B (en) | Pre-loan approval early warning method and system based on loan multi-head data | |
JP7017149B2 (en) | Information processing equipment, information processing method and information processing program using deep learning | |
CN107437227A (en) | Stock investment analysis apparatus and method | |
CN113095927B (en) | Method and equipment for identifying suspected transactions of backwashing money | |
Aksoy et al. | Artificial intelligence in computer-aided auditing techniques and technologies (CAATTs) and an application proposal for auditors | |
Chimonaki et al. | Identification of financial statement fraud in Greece by using computational intelligence techniques | |
Gupta et al. | Feature selection for dimension reduction of financial data for detection of financial statement frauds in context to Indian companies | |
Tu et al. | Online auction segmentation and effective selling strategy: Trust and information asymmetry perspectives | |
TWM587310U (en) | Screening system for potential buyers of financial commodities | |
CN113506173A (en) | Credit risk assessment method and related equipment thereof | |
TWM613536U (en) | Investment risk scoring system for fund commodities | |
CN117114812A (en) | Financial product recommendation method and device for enterprises | |
Oleksy et al. | On fine wine pricing across different trading venues | |
TWI769385B (en) | Method and system for screening potential purchasers of financial products | |
KR20240002900A (en) | Method for valuation of financial data on data economy and apparatus for performing the method | |
KR102000519B1 (en) | A method of recommending stock items using investment patterns of clients having similar tendency and system thereof | |
Badwan et al. | Financial Risk Management in the Supply Chain Using Business Intelligence and Big Data | |
CN114266594A (en) | Big data analysis method based on southeast Asia cross-border e-commerce platform | |
JP2001022851A (en) | Method and system for scoring | |
Bakhshi et al. | Developing a hybrid approach to credit priority based on accounting variables (using analytical network process (ANP) and multi-criteria decision-making) | |
Ben Ami et al. | Event-based trading: Building superior trading strategies with state-of-the-art information extraction tools |