WO2018016317A1 - Method for calculating insurance premium in which big data is used - Google Patents
Method for calculating insurance premium in which big data is used Download PDFInfo
- Publication number
- WO2018016317A1 WO2018016317A1 PCT/JP2017/024598 JP2017024598W WO2018016317A1 WO 2018016317 A1 WO2018016317 A1 WO 2018016317A1 JP 2017024598 W JP2017024598 W JP 2017024598W WO 2018016317 A1 WO2018016317 A1 WO 2018016317A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- insurance
- information
- policyholder
- premium
- pricing
- Prior art date
Links
- 238000000034 method Methods 0.000 title claims abstract description 12
- 238000013178 mathematical model Methods 0.000 claims abstract description 16
- 238000004891 communication Methods 0.000 claims description 5
- 238000005516 engineering process Methods 0.000 abstract description 5
- 230000010365 information processing Effects 0.000 abstract description 4
- 206010022000 influenza Diseases 0.000 description 5
- 230000035622 drinking Effects 0.000 description 2
- 239000004973 liquid crystal related substance Substances 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 230000003542 behavioural effect Effects 0.000 description 1
- 230000036772 blood pressure Effects 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 238000005336 cracking Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000001932 seasonal effect Effects 0.000 description 1
- UFTFJSFQGQCHQW-UHFFFAOYSA-N triformin Chemical compound O=COCC(OC=O)COC=O UFTFJSFQGQCHQW-UHFFFAOYSA-N 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/08—Insurance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
Definitions
- the present invention relates to a premium pricing method for insurance products using big data for insurance premium pricing.
- information relating to the insured person's information and behavior patterns can be acquired through a smartphone application or the like from SNS login information represented by the following.
- SNS login information represented by the following.
- LinkedIn It is possible to obtain information about work and education.
- Google Account Information on Gmail usage history (received and sent message history) can be obtained.
- Instagram It is possible to get updated photos based on the user's interests. In recent years, it is assumed that most smartphone users are using a plurality of the services described above, and many of them use personal authentication, so it is considered reliable as data.
- Non-Patent Document 1 Non-Patent Document 1
- Non-Patent Document 2 Molester insurance, lawyer insurance, mobile insurance, Harley-Davidson insurance, health age insurance, etc.
- Selling completely new insurance products that meet your needs is good for the development of the market, but it seems that there are not many cases where appropriate pricing is done. Many of these are not limited to the attributes of the insured, and often have the same premium.
- due to the nature of paying in case of any damages such as insurance it is difficult for the private companies to provide for those that involve moral hazard risk, even if there is a need. Many are difficult.
- employment insurance unemployment insurance
- GPS information of policyholders can be acquired using a smartphone application, which enables daily behavior patterns to be known, and it is possible to consider incorporating this into a mathematical model used for insurance premium pricing.
- the operator server displays the optimal insurance price pricing result on the app based on the mathematical model and the policyholder information.
- Insurance policyholders use the app to upload identity verification documents, confirm important explanation documents, and select payment methods, and exchange insurance policies on the app.
- a mathematical model (such as a hidden Markov model) that estimates whether it can be explained and predicted by such variables is constructed, and its parameters are estimated.
- New information resulting from the download of the application is always uploaded to the operator server, and the mathematical model and its parameters are updated from time to time as new information is acquired from the operator server.
- All insurance contracts are made via smartphone apps.
- the policyholder can download the app and input / upload necessary information to immediately know the insurance premium of the desired insurance purchase.
- insurance premiums are fixed regardless of each person as before, but by substituting insurance dividends or discounting insurance premiums for the next fiscal year by updating individual or overall information, the insurance premiums are substantially cheaper. I do. (Previously, there was a mechanism for paying insurance dividends to the entire group when the performance of the insurance group was good, and substituting the insurance premium for the insurance group. It is characterized by a number of parameters, and theoretically, each insured pays different insurance dividends according to the individual parameters.)
- the premium pricing method described above may be applicable to a variety of existing and new insurance products.
- the application to “mobile insurance” which has recently been launched will be considered.
- the pricing method according to the present invention to such insurance products, for example, information including the following is acquired. ⁇ Age, gender, educational background, work history ⁇ Location of home and work, commuting route ⁇ GPS movement information (how much does it move on a daily basis) ⁇ How to spend holidays (posted information on SNS) ⁇ Frequency of drinking (estimated)
- the accident rate of liquid crystal cracking due to falling mobile devices may have a correlation with the job type. The longer the commute time, the higher the daily movement, the higher the accident rate. The higher the frequency of going to the pool or waterfront on holidays, the higher the equipment submergence accident rate. The higher the frequency of drinking, the higher the accident rate.
- the pricing method of the present invention it is considered that policyholders who have been required to pay a high premium even though the risk of accident occurrence is low can join the insurance with a fairer premium.
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Economics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- Finance (AREA)
- Physics & Mathematics (AREA)
- Accounting & Taxation (AREA)
- Technology Law (AREA)
- Development Economics (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- Primary Health Care (AREA)
- Tourism & Hospitality (AREA)
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
[Problem] With regard to conventional pricing of an insurance product and a premium therefor, a statistically significant difference in an expected accident occurrence rate, if any, is reflected in the insurance premium, on the basis of, for life insurance, age and sex, and for property insurance, e.g., automobile insurance, information pertaining to the age and occurrence history of past similar accidents of a driver. However, although it has become possible to acquire various data different from those conventionally available, i.e., big data, due to recent progress in information processing technology, there are no products as yet in the insurance industry for which big data is utilized and premium pricing based on a mathematical model is practiced. Thus, the present state of affairs is such that a policyholder pays a high premium although he could otherwise be insured at a lower premium in accordance with risk. [Solution] The present invention provides a method for pricing the premium of an insurance product in which big data, which has become available to acquire and analyze due to recent information processing technology, is used for premium pricing.
Description
本発明は、ビッグデータを保険料価格付けに使用した保険商品の保険料価格付け方法に関する。
The present invention relates to a premium pricing method for insurance products using big data for insurance premium pricing.
The present invention relates to a premium pricing method for insurance products using big data for insurance premium pricing.
以下に代表されるようなSNSのログイン情報から、当該被保険者の情報及び行動パターン等に関する情報を、スマートフォンアプリ等を通じて取得可能であることは一般に知られている。
・Facebook: 年齢性別、学歴、余暇の過ごし方、余裕資金の使い方、位置情報、どのような記事やコメントを“いいね”と考えるかなどの情報取得が可能。どのような特性をもった人間と“友達”としてつながっているかとの情報取得が可能。
・Twitter: 日常の小さなつぶやきから行動パターンや思考パターンに関する情報取得が可能。
・LinkedIn: 職歴や学歴の情報取得が可能。
・Googleアカウント: Gmailの使用履歴（受信及び送信メッセージ履歴）の情報取得が可能。
・Instagram： 使用者の興味に基づきアップデートされた写真の取得が可能。
近年はほとんどのスマートフォン使用者は上記のうちの複数のサービス利用をしていることが想定され、またその多くは本人認証をしているため、データとして信頼のおけるものと考えられる。
It is generally known that information relating to the insured person's information and behavior patterns can be acquired through a smartphone application or the like from SNS login information represented by the following.
・ Facebook: You can get information such as age and gender, educational background, how to spend leisure time, how to use extra funds, location information, what kind of articles and comments you like. It is possible to obtain information on what characteristics people are connected to as “friends”.
・ Twitter: Information on behavioral patterns and thought patterns can be obtained from small daily tweets.
・ LinkedIn: It is possible to obtain information about work and education.
・ Google Account: Information on Gmail usage history (received and sent message history) can be obtained.
・ Instagram: It is possible to get updated photos based on the user's interests.
In recent years, it is assumed that most smartphone users are using a plurality of the services described above, and many of them use personal authentication, so it is considered reliable as data.
・Facebook: 年齢性別、学歴、余暇の過ごし方、余裕資金の使い方、位置情報、どのような記事やコメントを“いいね”と考えるかなどの情報取得が可能。どのような特性をもった人間と“友達”としてつながっているかとの情報取得が可能。
・Twitter: 日常の小さなつぶやきから行動パターンや思考パターンに関する情報取得が可能。
・LinkedIn: 職歴や学歴の情報取得が可能。
・Googleアカウント: Gmailの使用履歴（受信及び送信メッセージ履歴）の情報取得が可能。
・Instagram： 使用者の興味に基づきアップデートされた写真の取得が可能。
近年はほとんどのスマートフォン使用者は上記のうちの複数のサービス利用をしていることが想定され、またその多くは本人認証をしているため、データとして信頼のおけるものと考えられる。
It is generally known that information relating to the insured person's information and behavior patterns can be acquired through a smartphone application or the like from SNS login information represented by the following.
・ Facebook: You can get information such as age and gender, educational background, how to spend leisure time, how to use extra funds, location information, what kind of articles and comments you like. It is possible to obtain information on what characteristics people are connected to as “friends”.
・ Twitter: Information on behavioral patterns and thought patterns can be obtained from small daily tweets.
・ LinkedIn: It is possible to obtain information about work and education.
・ Google Account: Information on Gmail usage history (received and sent message history) can be obtained.
・ Instagram: It is possible to get updated photos based on the user's interests.
In recent years, it is assumed that most smartphone users are using a plurality of the services described above, and many of them use personal authentication, so it is considered reliable as data.
２００９年Google社では、新型インフルエンザが米国で流行することを数億にもなる検索キーワードと過去の季節性インフルエンザの流行との相関関係をビッグデータから分析し、特定の検索語４５個とある数式モデルの組み合わせから、その年のインフルエンザの流行を予測するモデルを構築した。これによりGoogle社は特定地域もしくは州単位での流行まで特定することができた。
このようなことが可能であったのは、Google社にインフルエンザのエキスパートがいたからではなく、大量のデータを処理し、統計的にどのような検索語が過去のインフルエンザ流行との相関が高いかを分析することができたからと考えられる。（非特許文献１）。
In 2009, Google analyzed the correlation between hundreds of millions of search keywords that caused the new influenza epidemic in the United States and the past seasonal influenza epidemic from big data. From the combination of models, we built a model to predict the influenza epidemic of the year. This allowed Google to identify the epidemic in a specific region or state.
This was not possible because Google had an influenza expert, but it processed a large amount of data and statistically found what search terms were highly correlated with past influenza outbreaks. It is thought that it was possible to analyze. (Non-Patent Document 1).
このようなことが可能であったのは、Google社にインフルエンザのエキスパートがいたからではなく、大量のデータを処理し、統計的にどのような検索語が過去のインフルエンザ流行との相関が高いかを分析することができたからと考えられる。（非特許文献１）。
In 2009, Google analyzed the correlation between hundreds of millions of search keywords that caused the new influenza epidemic in the United States and the past seasonal influenza epidemic from big data. From the combination of models, we built a model to predict the influenza epidemic of the year. This allowed Google to identify the epidemic in a specific region or state.
This was not possible because Google had an influenza expert, but it processed a large amount of data and statistically found what search terms were highly correlated with past influenza outbreaks. It is thought that it was possible to analyze. (Non-Patent Document 1).
本邦においては、[2006]年から小額短期保険制度が発足し、様々なアイデアをベースにした新しい保険商品が販売されてきている。（非特許文献２:例えば、痴漢保険、弁護士費用保険、モバイル保険、ハーレーダビッドソン保険、健康年齢保険など）
ニーズに合った全く新しい保険商品を販売することは市場の発展にとって好ましいことであるが、適切な価格付けが行なわれている事例は多くないと考えられる。これらの多くは被保険者の属性に限らず同一の保険料としているケースが多い。
また、保険という何らかの損害が出た際に支払いが発生するという性格上、モラルハザードのリスクが伴うものについて、例えニーズがあったとしても価格付けが困難であるため、私企業が提供することが困難なものも多い。例えば雇用保険（失業保険）は保険契約者が解雇されるリスクや求職しても就業できないリスクをカバーする性質上、従来の保険商品の枠組みでは価格付けが困難である。
In Japan, a small short-term insurance system has been established since [2006], and new insurance products based on various ideas have been sold. (Non-Patent Document 2: Molester insurance, lawyer insurance, mobile insurance, Harley-Davidson insurance, health age insurance, etc.)
Selling completely new insurance products that meet your needs is good for the development of the market, but it seems that there are not many cases where appropriate pricing is done. Many of these are not limited to the attributes of the insured, and often have the same premium.
In addition, due to the nature of paying in case of any damages such as insurance, it is difficult for the private companies to provide for those that involve moral hazard risk, even if there is a need. Many are difficult. For example, employment insurance (unemployment insurance) is difficult to price under the framework of conventional insurance products because it covers the risk of policyholders being dismissed and the risk of not being able to work even when seeking employment.
ニーズに合った全く新しい保険商品を販売することは市場の発展にとって好ましいことであるが、適切な価格付けが行なわれている事例は多くないと考えられる。これらの多くは被保険者の属性に限らず同一の保険料としているケースが多い。
また、保険という何らかの損害が出た際に支払いが発生するという性格上、モラルハザードのリスクが伴うものについて、例えニーズがあったとしても価格付けが困難であるため、私企業が提供することが困難なものも多い。例えば雇用保険（失業保険）は保険契約者が解雇されるリスクや求職しても就業できないリスクをカバーする性質上、従来の保険商品の枠組みでは価格付けが困難である。
In Japan, a small short-term insurance system has been established since [2006], and new insurance products based on various ideas have been sold. (Non-Patent Document 2: Molester insurance, lawyer insurance, mobile insurance, Harley-Davidson insurance, health age insurance, etc.)
Selling completely new insurance products that meet your needs is good for the development of the market, but it seems that there are not many cases where appropriate pricing is done. Many of these are not limited to the attributes of the insured, and often have the same premium.
In addition, due to the nature of paying in case of any damages such as insurance, it is difficult for the private companies to provide for those that involve moral hazard risk, even if there is a need. Many are difficult. For example, employment insurance (unemployment insurance) is difficult to price under the framework of conventional insurance products because it covers the risk of policyholders being dismissed and the risk of not being able to work even when seeking employment.
一方で、ビッグデータを活用し、保険料価格付けを行っている新しい（医療）保険商品の例も存在する。健康年齢少額短期保険（2016年6月17日商品発売開始）では、被保険者の年齢性別と身長体重に加え、一般的な健康診断で計測される最高最低血圧や中性脂肪値などの数値からウェブ上で健康年齢を算出し、その健康年齢に従って保険料を決定している。健康年齢少額短期保険では、グループ会社に株式会社日本医療情報センター（JMDC）があり、１６０万件超の匿名化されたレセプト（診療報酬明細書）情報のビッグデータを基に保険料の価格付けを行っている。（非特許文献３）
On the other hand, there are also examples of new (medical) insurance products that use big data and price premiums. In the short-term health age short-term insurance (launched on June 17, 2016), in addition to the insured's age and gender and height, numerical values such as the highest and lowest blood pressure and triglyceride levels measured by general health examination The health age is calculated on the web and the insurance premium is determined according to the health age. For health age small-amount short-term insurance, the group company has the Japan Medical Information Center Co., Ltd. (JMDC). Pricing of insurance premiums is based on the big data of more than 1.6 million anonymized receipts. It is carried out. (Non Patent Literature 3)
On the other hand, there are also examples of new (medical) insurance products that use big data and price premiums. In the short-term health age short-term insurance (launched on June 17, 2016), in addition to the insured's age and gender and height, numerical values such as the highest and lowest blood pressure and triglyceride levels measured by general health examination The health age is calculated on the web and the insurance premium is determined according to the health age. For health age small-amount short-term insurance, the group company has the Japan Medical Information Center Co., Ltd. (JMDC). Pricing of insurance premiums is based on the big data of more than 1.6 million anonymized receipts. It is carried out. (Non Patent Literature 3)
SNS以外にも、例えば、クレジットカードの明細や銀行口座の明細の管理サービスを行っている事業者（非特許文献４）との提携や、当該事業者からのデータ購入により、収入・購買履歴等を把握し、これを保険料価格付けに用いる数理モデルに組み入れることも検討可能。
Other than SNS, for example, income / purchase history, etc. due to tie-ups with businesses (non-patent literature 4) that provide credit card details and bank account details management services, and data purchases from such businesses It is also possible to consider this and incorporate it into a mathematical model used for premium pricing.
Other than SNS, for example, income / purchase history, etc. due to tie-ups with businesses (non-patent literature 4) that provide credit card details and bank account details management services, and data purchases from such businesses It is also possible to consider this and incorporate it into a mathematical model used for premium pricing.
更に、スマートフォンアプリにより、保険契約者のGPS情報も取得可能であり、これにより日常の行動パターンを知ることが可能であり、これを保険料価格付けに用いる数理モデルに組み入れることも検討可能。
Furthermore, GPS information of policyholders can be acquired using a smartphone application, which enables daily behavior patterns to be known, and it is possible to consider incorporating this into a mathematical model used for insurance premium pricing.
Furthermore, GPS information of policyholders can be acquired using a smartphone application, which enables daily behavior patterns to be known, and it is possible to consider incorporating this into a mathematical model used for insurance premium pricing.
保険分野ではないが、香港のWeLab社では、オンラインでの個人への貸付を行っており、その際Facebook情報の提供で貸付金利を５％オフ、LinkedIn情報の提供で貸付金利を１０％オフしているとのこと。更に、何時にどの程度の時間電話で通話するかということ（例えば、夜、長電話をする人は貸し倒れ確率が高いもしくは低い、など）も貸付金利を決定するための変数としているとのこと。
（非特許文献５）
Although not in the insurance field, WeLab in Hong Kong provides loans to individuals online, providing Facebook information with a 5% off loan interest rate and LinkedIn information providing a 10% off loan interest rate. It is said that. In addition, what time and how long the telephone call is made (for example, the person who makes a long call at night has a high or low probability of credit loss) is also a variable for determining the loan interest rate.
(Non-Patent Document 5)
（非特許文献５）
Although not in the insurance field, WeLab in Hong Kong provides loans to individuals online, providing Facebook information with a 5% off loan interest rate and LinkedIn information providing a 10% off loan interest rate. It is said that. In addition, what time and how long the telephone call is made (for example, the person who makes a long call at night has a high or low probability of credit loss) is also a variable for determining the loan interest rate.
(Non-Patent Document 5)
しかしながら、依然として保険業界では上記のような情報処理技術によるビッグデータを使用した保険料価格付けがなされていない、もしくはそのような技術を使用しないため潜在的ニーズがあるにも関わらず提供できない保険商品が存在しているのが現状である。
However, the insurance industry still does not charge premiums using big data based on information processing technology as described above, or insurance products that cannot be provided despite the potential needs because such technology is not used. Is present.
However, the insurance industry still does not charge premiums using big data based on information processing technology as described above, or insurance products that cannot be provided despite the potential needs because such technology is not used. Is present.
以下では、通信情報端末の一例として、スマートフォンを使用し、特に保険事業者が無料提供するスマートフォンアプリにより保険料価格付け及び保険販売を行うことを想定する。
In the following, it is assumed that a smartphone is used as an example of a communication information terminal, and insurance premium pricing and insurance sales are performed by a smartphone application provided free of charge by an insurance company.
In the following, it is assumed that a smartphone is used as an example of a communication information terminal, and insurance premium pricing and insurance sales are performed by a smartphone application provided free of charge by an insurance company.
・保険事業者はスマートフォンアプリを開発し、無償でダウンロードできるようにする。（保険契約のためのウェブサイトを公開しそこでデータ提供および保険契約を行うことが出来るようにする）
・保険契約者は、自己に適用される保険料の水準を知るためには、ＳＮＳのログイン情報をアプリに入力することで保険事業者に提供する必要があるものとする。これにより保険事業者は、SNS等サーバーに存在する保険契約者情報を取得することが可能となる。
・保険事業者は、このような情報を格納するサーバー（事業者サーバー）を用意する。なお、情報は（潜在）保険契約者以外にも、提携事業者等から提供を受けることも考えられる。
・保険事業者は、事業者サーバーに存在するデータから、数理モデルのパラメータを推定し、与えられた情報と保険事故が将来発生する相関関係を推定するモデル設定する。
・保険契約者が必要な情報をアプリに入力すると、事業者サーバーは数理モデルと保険契約者情報を基に、最適な保険料の価格付け結果をアプリで表示する。
・保険契約者は、本人確認書類アップロードや重要事項説明書類の確認、決済方法の選択をアプリで行い、保険契約をアプリ上で取り交わす。
・ Insurers will develop smartphone apps that can be downloaded free of charge. (A website for insurance contracts will be published so that data can be provided and insurance contracts can be made there)
-In order for the policyholder to know the level of the premium applied to himself / herself, it is necessary to provide the insurance company with the SNS login information entered in the app. As a result, the insurance company can acquire the policyholder information existing in the server such as SNS.
-The insurance company prepares a server (operator server) for storing such information. In addition to (potential) policyholders, the information may be provided by partner companies.
The insurance company estimates a mathematical model parameter from data existing in the company server, and sets a model for estimating the correlation between the given information and an insurance accident in the future.
・ When the policyholder inputs the necessary information into the app, the operator server displays the optimal insurance price pricing result on the app based on the mathematical model and the policyholder information.
・ Insurance policyholders use the app to upload identity verification documents, confirm important explanation documents, and select payment methods, and exchange insurance policies on the app.
・保険契約者は、自己に適用される保険料の水準を知るためには、ＳＮＳのログイン情報をアプリに入力することで保険事業者に提供する必要があるものとする。これにより保険事業者は、SNS等サーバーに存在する保険契約者情報を取得することが可能となる。
・保険事業者は、このような情報を格納するサーバー（事業者サーバー）を用意する。なお、情報は（潜在）保険契約者以外にも、提携事業者等から提供を受けることも考えられる。
・保険事業者は、事業者サーバーに存在するデータから、数理モデルのパラメータを推定し、与えられた情報と保険事故が将来発生する相関関係を推定するモデル設定する。
・保険契約者が必要な情報をアプリに入力すると、事業者サーバーは数理モデルと保険契約者情報を基に、最適な保険料の価格付け結果をアプリで表示する。
・保険契約者は、本人確認書類アップロードや重要事項説明書類の確認、決済方法の選択をアプリで行い、保険契約をアプリ上で取り交わす。
・ Insurers will develop smartphone apps that can be downloaded free of charge. (A website for insurance contracts will be published so that data can be provided and insurance contracts can be made there)
-In order for the policyholder to know the level of the premium applied to himself / herself, it is necessary to provide the insurance company with the SNS login information entered in the app. As a result, the insurance company can acquire the policyholder information existing in the server such as SNS.
-The insurance company prepares a server (operator server) for storing such information. In addition to (potential) policyholders, the information may be provided by partner companies.
The insurance company estimates a mathematical model parameter from data existing in the company server, and sets a model for estimating the correlation between the given information and an insurance accident in the future.
・ When the policyholder inputs the necessary information into the app, the operator server displays the optimal insurance price pricing result on the app based on the mathematical model and the policyholder information.
・ Insurance policyholders use the app to upload identity verification documents, confirm important explanation documents, and select payment methods, and exchange insurance policies on the app.
本発明によれば、情報処理技術により取得分析可能となったビッグデータを活用した保険料の価格付けを行うことが可能となり、より合理的な価格で手軽に保険に加入することが可能となる。
According to the present invention, it is possible to price insurance premiums utilizing big data that can be acquired and analyzed by information processing technology, and it is possible to easily purchase insurance at a more reasonable price. .
According to the present invention, it is possible to price insurance premiums utilizing big data that can be acquired and analyzed by information processing technology, and it is possible to easily purchase insurance at a more reasonable price. .
スマートフォンアプリを製作し、当該アプリにより、
・各種SNSへのログイン情報の取得
・保険契約およびそれに必要な重要事項説明、本人確認書類のアップロード
・保険料決済方法（クレジットカードなど）の選択
・事業者への意見質問等の発信
を行うことが出来るようにする。当該アプリは無料でアップルストア等からダウンロードできるものとする。
Make a smartphone app,
・ Acquisition of login information to various SNS ・ Insurance contract and explanation of important matters necessary for it, upload of identification documents ・ Selection of insurance payment method (credit card, etc.) ・ Discussion of opinions and questions to businesses To be able to The app can be downloaded free of charge from an Apple store or the like.
・各種SNSへのログイン情報の取得
・保険契約およびそれに必要な重要事項説明、本人確認書類のアップロード
・保険料決済方法（クレジットカードなど）の選択
・事業者への意見質問等の発信
を行うことが出来るようにする。当該アプリは無料でアップルストア等からダウンロードできるものとする。
Make a smartphone app,
・ Acquisition of login information to various SNS ・ Insurance contract and explanation of important matters necessary for it, upload of identification documents ・ Selection of insurance payment method (credit card, etc.) ・ Discussion of opinions and questions to businesses To be able to The app can be downloaded free of charge from an Apple store or the like.
別途各種SNSの情報を用意（購入もしくは他の事業者との提携により入手、さらに当該アプリのダウンロード数が増加することで取得情報は自然に増加する）し、開発する保険商品の保険事故がどのような変数によって説明・予測が可能かを推定する数理モデル（例えば隠れマルコフモデル等）を構築し、そのパラメータ推計を行う。
Separately prepare various SNS information (obtained by purchasing or tie-up with other companies, and further increase the number of downloads of the app, the acquired information will naturally increase) A mathematical model (such as a hidden Markov model) that estimates whether it can be explained and predicted by such variables is constructed, and its parameters are estimated.
Separately prepare various SNS information (obtained by purchasing or tie-up with other companies, and further increase the number of downloads of the app, the acquired information will naturally increase) A mathematical model (such as a hidden Markov model) that estimates whether it can be explained and predicted by such variables is constructed, and its parameters are estimated.
当該アプリのダウンロードによる新規情報は常に事業者サーバーにアップロードされ、事業者サーバーの新規情報取得に伴い上記数理モデル及びそのパラメータは随時アップデートされる。
New information resulting from the download of the application is always uploaded to the operator server, and the mathematical model and its parameters are updated from time to time as new information is acquired from the operator server.
New information resulting from the download of the application is always uploaded to the operator server, and the mathematical model and its parameters are updated from time to time as new information is acquired from the operator server.
保険契約は全てスマートフォンアプリを経由して契約が行われる。保険契約者はアプリをダウンロードし、必要な情報を入力・アップロードすることにより購入希望する保険の保険料を即座に知ることが可能となる。
All insurance contracts are made via smartphone apps. The policyholder can download the app and input / upload necessary information to immediately know the insurance premium of the desired insurance purchase.
All insurance contracts are made via smartphone apps. The policyholder can download the app and input / upload necessary information to immediately know the insurance premium of the desired insurance purchase.
これらの仕組み（アプリ、事業者サーバー上の情報、数理モデル）により、保険料の価格付けを行い、（安く提供可能な被保険者には）安い保険料を提供可能とする。
With these mechanisms (applications, information on company servers, mathematical models), insurance premiums are priced, and cheaper insurance premiums can be provided (for insured persons who can offer them cheaply).
With these mechanisms (applications, information on company servers, mathematical models), insurance premiums are priced, and cheaper insurance premiums can be provided (for insured persons who can offer them cheaply).
もしくは、保険料は従来どおり各人によらず一定とするが、個別もしくは全体の情報更新により、保険配当を行ったり来年度の保険料割引を行ったりすることで、実質的に安い保険料価格付けを行う。（従来、ある保険群団の成績が良かった場合にその群団全体に保険配当を行い実質的な保険料割引をする仕組みは存在していた。しかしここでの保険配当は、被保険者を多数のパラメータで特徴づけを行い、その個別のパラメータに従い理論的には各被保険者ごとに異なる保険配当と行うものである。）
Or, insurance premiums are fixed regardless of each person as before, but by substituting insurance dividends or discounting insurance premiums for the next fiscal year by updating individual or overall information, the insurance premiums are substantially cheaper. I do. (Previously, there was a mechanism for paying insurance dividends to the entire group when the performance of the insurance group was good, and substituting the insurance premium for the insurance group. It is characterized by a number of parameters, and theoretically, each insured pays different insurance dividends according to the individual parameters.)
Or, insurance premiums are fixed regardless of each person as before, but by substituting insurance dividends or discounting insurance premiums for the next fiscal year by updating individual or overall information, the insurance premiums are substantially cheaper. I do. (Previously, there was a mechanism for paying insurance dividends to the entire group when the performance of the insurance group was good, and substituting the insurance premium for the insurance group. It is characterized by a number of parameters, and theoretically, each insured pays different insurance dividends according to the individual parameters.)
逆に、情報の更新によりリスク顕在化した際には、保険料の増加もしくは保険金額の減少を行う。
On the contrary, when the risk becomes obvious by updating the information, the insurance premium is increased or the insurance amount is decreased.
On the contrary, when the risk becomes obvious by updating the information, the insurance premium is increased or the insurance amount is decreased.
上記のような保険料の価格付けの手法は、様々な既存の保険商品及び新しい保険商品に適用可能と考えられる。以下では、一例として近年販売開始された“モバイル保険“への応用を検討する。
The premium pricing method described above may be applicable to a variety of existing and new insurance products. In the following, as an example, the application to “mobile insurance” which has recently been launched will be considered.
The premium pricing method described above may be applicable to a variety of existing and new insurance products. In the following, as an example, the application to “mobile insurance” which has recently been launched will be considered.
さくら小額短期保険株式会社により2016年5月13日より販売開始されている。対象のモバイル機器（例えばスマートフォン）の液晶割れや水没事故に対して一定の条件下で修理費用（上限あり）を支払う保険である。月額保険料は一律700円であり、過去の同様の事象の事故率などの経験をもとに価格付けをしていると思われる。
It has been on sale since May 13, 2016 by Sakura Small Value Insurance Co., Ltd. This insurance pays repair costs (with an upper limit) under certain conditions against liquid crystal breakage or submersion accidents on the target mobile device (for example, a smartphone). The monthly premium is 700 yen, which seems to be priced based on experience such as the accident rate of similar events in the past.
It has been on sale since May 13, 2016 by Sakura Small Value Insurance Co., Ltd. This insurance pays repair costs (with an upper limit) under certain conditions against liquid crystal breakage or submersion accidents on the target mobile device (for example, a smartphone). The monthly premium is 700 yen, which seems to be priced based on experience such as the accident rate of similar events in the past.
このような保険商品に対して、本発明による価格付け方法を適用するため、例えば、以下を含む情報を取得する。
・年齢、性別、学歴、職歴
・自宅と勤務先の位置、通勤経路
・GPSによる移動情報（１日にどの程度日常的に移動するか）
・休日の過ごし方（SNSへの投稿情報）
・飲酒の頻度（推定）
例えば、モバイル機器の落下による液晶割れ等の事故率は、職種との相関関係があるかもしれない。通勤時間が長ければ長いほど、１日の移動量が多ければ多いほど事故率が高いかもしれない。休日にプールや水辺にいく頻度が高いほど機器お水没事故率が高いかもしれない。飲酒頻度が高いほど事故率が高いかもしれない。本発明の価格付け方法により、事故発生リスクが低いと考えられるにもかかわらず高い保険料を払う必要があった保険契約者がより公平な保険料で保険加入が可能となると考えられる。
In order to apply the pricing method according to the present invention to such insurance products, for example, information including the following is acquired.
・ Age, gender, educational background, work history ・ Location of home and work, commuting route ・ GPS movement information (how much does it move on a daily basis)
・ How to spend holidays (posted information on SNS)
・ Frequency of drinking (estimated)
For example, the accident rate of liquid crystal cracking due to falling mobile devices may have a correlation with the job type. The longer the commute time, the higher the daily movement, the higher the accident rate. The higher the frequency of going to the pool or waterfront on holidays, the higher the equipment submergence accident rate. The higher the frequency of drinking, the higher the accident rate. According to the pricing method of the present invention, it is considered that policyholders who have been required to pay a high premium even though the risk of accident occurrence is low can join the insurance with a fairer premium.
・年齢、性別、学歴、職歴
・自宅と勤務先の位置、通勤経路
・GPSによる移動情報（１日にどの程度日常的に移動するか）
・休日の過ごし方（SNSへの投稿情報）
・飲酒の頻度（推定）
例えば、モバイル機器の落下による液晶割れ等の事故率は、職種との相関関係があるかもしれない。通勤時間が長ければ長いほど、１日の移動量が多ければ多いほど事故率が高いかもしれない。休日にプールや水辺にいく頻度が高いほど機器お水没事故率が高いかもしれない。飲酒頻度が高いほど事故率が高いかもしれない。本発明の価格付け方法により、事故発生リスクが低いと考えられるにもかかわらず高い保険料を払う必要があった保険契約者がより公平な保険料で保険加入が可能となると考えられる。
In order to apply the pricing method according to the present invention to such insurance products, for example, information including the following is acquired.
・ Age, gender, educational background, work history ・ Location of home and work, commuting route ・ GPS movement information (how much does it move on a daily basis)
・ How to spend holidays (posted information on SNS)
・ Frequency of drinking (estimated)
For example, the accident rate of liquid crystal cracking due to falling mobile devices may have a correlation with the job type. The longer the commute time, the higher the daily movement, the higher the accident rate. The higher the frequency of going to the pool or waterfront on holidays, the higher the equipment submergence accident rate. The higher the frequency of drinking, the higher the accident rate. According to the pricing method of the present invention, it is considered that policyholders who have been required to pay a high premium even though the risk of accident occurrence is low can join the insurance with a fairer premium.
Claims (2)
- 通信情報端末において保険を販売する方法において、
保険事業者と、保険契約者と、ＳＮＳサーバーと、事業者サーバーと、事業者サーバーにある情報に基づいて推定されるモデルパラメータを持つ数理モデルと、
に関して、
保険契約者が通信情報端末でＳＮＳログイン情報及びクレジットカード購買履歴や銀行口座取引履歴等の保険契約者に関する情報を入力し保険事業者に送付するステップ、
保険事業者が当該ＳＮＳログイン情報から得られる前記保険契約者に関する情報を事業者サーバーに格納するステップ、
アップデートされたサーバー情報に基づいて数理モデル及び数理モデルパラメータを推定するステップ、
前記保険契約者の情報を数理モデルに入力することで保険料価格付けを行い、保険料を保険契約者に提示するステップ、
とからなる保険商品の価格付け及び販売方法。
In a method of selling insurance at a communication information terminal,
An insurer, a policyholder, an SNS server, an operator server, a mathematical model with model parameters estimated based on information in the operator server;
With respect to
The policyholder inputs information related to the policyholder such as SNS login information and credit card purchase history and bank account transaction history at the communication information terminal, and sends the information to the insurance operator.
Storing information on the policyholder obtained by the insurance company from the SNS login information in the company server;
Estimating a mathematical model and mathematical model parameters based on the updated server information;
Performing premium pricing by inputting the policyholder information into a mathematical model and presenting the premium to the policyholder;
Pricing and sales methods for insurance products consisting of
- 通信情報端末において保険を販売する場合の保険料の価格付け方法において、
保険事業者と、保険契約者と、ＳＮＳサーバーと、事業者サーバーと、事業者サーバーにある情報に基づいて推定されるモデルパラメータを持つ数理モデルと、
に関して、
保険契約者が通信情報端末でＳＮＳログイン情報及びクレジットカード購買履歴や銀行口座取引履歴やGPS情報を含む移動情報等の保険契約者に関する情報を入力し保険事業者に送付するステップ、
保険事業者が当該ＳＮＳログイン情報から得られる前記保険契約者に関する情報を事業者サーバーに格納するステップ、
アップデートされたサーバー情報に基づいて数理モデル及び数理モデルパラメータを推定するステップ、
保険料は各契約者同一とするが、前記保険契約者の情報を数理モデルに入力することで契約者別に保険料を事後的に価格付けするステップ、
とからなる保険商品の価格付け方法。
In the insurance premium pricing method when selling insurance on communication information terminals,
An insurer, a policyholder, an SNS server, an operator server, a mathematical model with model parameters estimated based on information in the operator server;
With respect to
The policyholder inputs information related to the policyholder such as SNS login information, credit card purchase history, bank account transaction history, and movement information including GPS information at the communication information terminal and sends it to the insurance operator;
Storing information on the policyholder obtained by the insurance company from the SNS login information in the company server;
Estimating a mathematical model and mathematical model parameters based on the updated server information;
The insurance premiums are the same for each policyholder, and the premiums are subsequently priced by policyholder by inputting the policyholder information into the mathematical model,
A method for pricing insurance products.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP2016-142986 | 2016-07-20 | ||
| JP2016142986A JP6145724B1 (en) | 2016-07-21 | 2016-07-21 | How to calculate insurance premiums using big data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2018016317A1 true WO2018016317A1 (en) | 2018-01-25 |
Family
ID=59061147
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/JP2017/024598 WO2018016317A1 (en) | 2016-07-20 | 2017-07-05 | Method for calculating insurance premium in which big data is used |
Country Status (2)
| Country | Link |
|---|---|
| JP (1) | JP6145724B1 (en) |
| WO (1) | WO2018016317A1 (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110009507A (en) * | 2018-11-27 | 2019-07-12 | 阿里巴巴集团控股有限公司 | A kind of insurance transaction processing method and device |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2002373259A (en) * | 2001-03-29 | 2002-12-26 | Mizuho Dl Financial Technology Co Ltd | Net premium calculation method in property insurance or the like using individual risk model and system therefor |
| JP2003006439A (en) * | 2001-06-26 | 2003-01-10 | Toshiba Corp | Method and system for damage insurance data acquisition, program for damage insurance data, and program for providing damage insurance data |
| JP2006139724A (en) * | 2004-11-15 | 2006-06-01 | Junji Uchida | Insurance business operation system |
| WO2006082957A1 (en) * | 2005-02-07 | 2006-08-10 | Kumazawa, Noboru | Longevity insurance system and its method |
| JP2011070581A (en) * | 2009-09-28 | 2011-04-07 | Csk Corp | Insurance system and insurance program |
| US20120221357A1 (en) * | 2011-02-28 | 2012-08-30 | Krause Jacqueline Lesage | Systems and methods for intelligent underwriting based on community or social network data |
| WO2014181530A1 (en) * | 2013-05-08 | 2014-11-13 | パナソニック インテレクチュアル プロパティ コーポレーション オブ アメリカ | Service provision method |
| JP2015138406A (en) * | 2014-01-22 | 2015-07-30 | 日本Ｒａ株式会社 | Vehicle insurance system and vehicle insurance presenting method |
-
2016
- 2016-07-21 JP JP2016142986A patent/JP6145724B1/en not_active Expired - Fee Related
-
2017
- 2017-07-05 WO PCT/JP2017/024598 patent/WO2018016317A1/en active Application Filing
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2002373259A (en) * | 2001-03-29 | 2002-12-26 | Mizuho Dl Financial Technology Co Ltd | Net premium calculation method in property insurance or the like using individual risk model and system therefor |
| JP2003006439A (en) * | 2001-06-26 | 2003-01-10 | Toshiba Corp | Method and system for damage insurance data acquisition, program for damage insurance data, and program for providing damage insurance data |
| JP2006139724A (en) * | 2004-11-15 | 2006-06-01 | Junji Uchida | Insurance business operation system |
| WO2006082957A1 (en) * | 2005-02-07 | 2006-08-10 | Kumazawa, Noboru | Longevity insurance system and its method |
| JP2011070581A (en) * | 2009-09-28 | 2011-04-07 | Csk Corp | Insurance system and insurance program |
| US20120221357A1 (en) * | 2011-02-28 | 2012-08-30 | Krause Jacqueline Lesage | Systems and methods for intelligent underwriting based on community or social network data |
| WO2014181530A1 (en) * | 2013-05-08 | 2014-11-13 | パナソニック インテレクチュアル プロパティ コーポレーション オブ アメリカ | Service provision method |
| JP2015138406A (en) * | 2014-01-22 | 2015-07-30 | 日本Ｒａ株式会社 | Vehicle insurance system and vehicle insurance presenting method |
Also Published As
| Publication number | Publication date |
|---|---|
| JP6145724B1 (en) | 2017-06-14 |
| JP2018013953A (en) | 2018-01-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9431028B2 (en) | Apparatuses, methods and systems for a digital conversation management platform | |
| CA2894101A1 (en) | A financial product recommendation for a consumer | |
| US20160232546A1 (en) | Computer processing of financial product information and information about consumers of financial products | |
| US20160092993A1 (en) | Computer readable medium, system, and method of providing a virtual venue for the transfer of taxpayer-specific information | |
| US11556996B1 (en) | Method and system for integrating online sales of consumer products with the sale of corresponding insurance | |
| WO2014121002A1 (en) | Business financing method and system | |
| WO2017034643A1 (en) | Systems and methods for processing charges for disputed transactions | |
| US20130204701A1 (en) | Apparatus, system and methods for marketing targeted products to users of social media | |
| WO2019216825A1 (en) | System and method for recommending insurance products | |
| US20230058474A1 (en) | Multi-modal routing engine and processing architecture for orchestration of multi-currency lending exchanges | |
| Mierzwinski et al. | Selling consumers not lists: The new world of digital decision-making and the role of the fair credit reporting act | |
| US20160239920A1 (en) | Crowd Funding Fraud Insurance | |
| US20240152998A1 (en) | Methods and systems for providing personalized purchasing information | |
| US10325252B2 (en) | Payment management apparatus, payment management method, and storage medium | |
| Fletcher | Disclosure as a tool for enhancing consumer engagement and competition | |
| Rabaa’i | FinTech in Kuwait: a survey study | |
| US20210182933A1 (en) | Methods and systems for providing a vehicle recommendation | |
| Ali et al. | An integrated framework for mobile payment in Pakistan: drivers, barriers, and facilitators of usage behavior | |
| US20130024386A1 (en) | Engine, system and method of providing business valuation and database services using alternative payment arrangments | |
| US11556836B1 (en) | System and method for matching specialists and potential clients | |
| US20180121944A1 (en) | User data currency normalization | |
| WO2018016317A1 (en) | Method for calculating insurance premium in which big data is used | |
| Kabulova | Impact of FinTech innovation on the financial sector’s stability | |
| US20170256009A1 (en) | Marketing real estate services based on listing status | |
| Meltzer | Using the Internet to promote services exports by small-and medium-sized enterprises |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 17830837 Country of ref document: EP Kind code of ref document: A1 |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| 32PN | Ep: public notification in the ep bulletin as address of the adressee cannot be established |
Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 21.05.2019) |
|
| NENP | Non-entry into the national phase |
Ref country code: JP |
|
| 122 | Ep: pct application non-entry in european phase |
Ref document number: 17830837 Country of ref document: EP Kind code of ref document: A1 |
|
| NENP | Non-entry into the national phase |
Ref country code: JP |
|
| NENP | Non-entry into the national phase |
Ref country code: JP |