WO2022239246A1 - Information processing device, information processing method, and program - Google Patents

Information processing device, information processing method, and program Download PDF

Info

Publication number
WO2022239246A1
WO2022239246A1 PCT/JP2021/018489 JP2021018489W WO2022239246A1 WO 2022239246 A1 WO2022239246 A1 WO 2022239246A1 JP 2021018489 W JP2021018489 W JP 2021018489W WO 2022239246 A1 WO2022239246 A1 WO 2022239246A1
Authority
WO
WIPO (PCT)
Prior art keywords
information
reservation
applicant
model
unit
Prior art date
Application number
PCT/JP2021/018489
Other languages
French (fr)
Japanese (ja)
Inventor
唯史 藤井
寛 吉田
朋子 柴田
Original Assignee
日本電信電話株式会社
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 日本電信電話株式会社 filed Critical 日本電信電話株式会社
Priority to PCT/JP2021/018489 priority Critical patent/WO2022239246A1/en
Publication of WO2022239246A1 publication Critical patent/WO2022239246A1/en

Links

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
    • G06Q10/00—Administration; Management
    • G06Q10/02—Reservations, e.g. for tickets, services or events

Definitions

  • the present invention relates to an information processing device, an information processing method, and a program.
  • Reservation systems are known for applying for reservations for services such as restaurants and lodging facilities from terminal devices owned by customers.
  • the service provider obtains attribute information indicating the customer's age, sex, address, and other attributes as information identifying the customer when applying for a reservation.
  • Patent Literature 1 discloses a technology that makes it possible to grasp a customer's attribute in a reservation system that uses a customer's telephone number as information that identifies the customer. The technology disclosed in Patent Literature 1 gives a benefit to a customer in return for the customer's provision of his/her own attribute information.
  • the purpose of the present invention is to provide technology for improving the quality of reservation operations.
  • An information processing apparatus includes: a first acquisition unit that acquires weight information indicating a weight for at least one characteristic of a reservation person; wherein the reservation status data relating to each of the plurality of reserving persons indicates the details of reservations completed for the reserving person, and characteristic information indicating characteristics of the reserving person. and an extraction unit, using the extracted reservation status data as learning data, inputting characteristic information indicating characteristics of the applicant and candidate date information indicating a candidate date for service implementation, and inputting the candidate date for each, a learning unit that trains a model that outputs a score representing the applicant's expected satisfaction with the service being performed on the candidate date.
  • FIG. 1 is a block diagram showing an information processing apparatus according to an embodiment of the invention.
  • FIG. 2 is a block diagram showing the configuration of the model generating unit shown in FIG. 1;
  • FIG. 3 is a diagram for explaining scores output from a model generated by the model generation unit shown in FIG.
  • FIG. 4 is a block diagram showing the configuration of the information presentation unit shown in FIG.
  • FIG. 5 is a block diagram showing the hardware configuration of the information processing apparatus shown in FIG. 1.
  • FIG. 6 is a flow chart showing the procedure of model generation according to the embodiment of the present invention.
  • FIG. 7 is a flow chart showing the information presentation procedure according to the embodiment of the present invention.
  • An embodiment of the present invention relates to an information processing apparatus applicable to a reservation system that accepts a reservation application for a service from a customer and makes a reservation according to the customer's request.
  • a customer who applies for a reservation is referred to as an applicant.
  • a customer who has completed a reservation is referred to as a reservation person.
  • the embodiment will be explained using a reservation system for making reservations for optical line construction as an example.
  • An applicant who wants to use the optical line internet service at home applies for the use of the optical line internet service to an internet service provider.
  • the presence of the applicant or a person concerned is required for the optical line construction.
  • the receptionist of the Internet service provider adjusts the schedule of the optical line construction with the applicant by telephone.
  • FIG. 1 schematically shows an information processing device 10 according to an embodiment of the invention.
  • the information processing device 10 includes a model generation unit 11, an information presentation unit 12, a reservation status storage unit 13, and a model storage unit .
  • the reservation status storage unit 13 stores reservation status data for each of a plurality of subscribers.
  • the reservation status data is data indicating the details of completed reservations (past reservations).
  • the reservation status data includes characteristic information and application information of the reservation person.
  • Characteristic information is information that indicates one or more characteristics of a customer (eg, applicant or reservation person). Examples of characteristics include age, gender, tone of voice (mood).
  • Voice tone is represented by two values, for example normal and displeased (angry). Note that the voice tone may be represented by three or more values.
  • the tone of voice may be evaluated by the receptionist who spoke with the person who made the reservation at the time of the reservation, or by analyzing the voice of the person who made the reservation at the time of the reservation with voice analysis software.
  • the application information includes, for example, information indicating an application service, information indicating an application time, and candidate date information indicating a candidate date for service implementation.
  • the application service is optical line construction.
  • the application time includes, for example, the application date when the reservation for the optical line construction was made and the construction date when the optical line construction is to be performed. Based on the information indicating the application period, it is possible to calculate the number of days from the application to the implementation of the optical line construction.
  • the candidate date information indicates the candidate date for optical line installation.
  • the candidate date for optical line installation indicates a free installation date, that is, a date on which the Internet service provider can perform the optical line installation.
  • the candidate date is represented by the number of days from the application date.
  • the application information includes the score for the construction date. Scores will be discussed later.
  • the model generation unit 11 uses the reservation status data stored in the reservation status storage unit 13 to predict the desired construction date, which is the date on which the applicant wishes to perform the optical line construction.
  • a model is generated and the generated machine learning model is stored in the model storage unit 14 .
  • the model storage unit 14 stores multiple models generated by the model generation unit 11 .
  • the information presentation unit 12 uses one of a plurality of models stored in the model storage unit 14 to predict the applicant's desired construction date based on the applicant information, and outputs the prediction result. do.
  • Applicant information includes applicant's characteristic information and information indicating an applied service.
  • the person in charge of reception can refer to the prediction result output by the information presentation unit 12 and propose the construction date to the applicant. After completing the reservation (schedule adjustment), the person in charge of reception registers the contents of the reservation in the reservation status storage unit 13 .
  • model generation unit 11 and the information presentation unit 12 exist in one device.
  • the model generator 11 and the information presenter 12 may exist in separate devices.
  • FIG. 2 schematically shows a configuration example of the model generation unit 11.
  • the model generation unit 11 includes an acquisition unit 111 , an extraction unit 112 and a learning unit 113 .
  • the acquisition unit 111 acquires weight information indicating a weight for at least one characteristic of the reservation person.
  • the weight information can be input by a user who manages the information processing device 10 .
  • Weight information may include conditions that specify values or ranges for each property.
  • the weight information may further include information specifying the subscription service.
  • the extraction unit 112 extracts reservation status data corresponding to the weight information acquired by the acquisition unit 111 from the reservation status data stored in the reservation status storage unit 13 .
  • the extraction unit 112 acquires reservation status data including characteristic information that matches the conditions included in the weight information from the reservation status storage unit 13 .
  • the weight information includes a condition specifying a displeased voice tone
  • the extraction unit 112 acquires reservation status data including characteristic information in which the value of the characteristic "voice tone" is "displeased” from the reservation status storage unit 13. .
  • the extraction unit 112 extracts from the reservation status storage unit 13 the value of the characteristic “age” of 20. 29, the value of the property “sex” is “male”, and the value of the property “voice tone” is “sullen”.
  • reservation status data includes offer information in addition to property information.
  • the learning unit 113 learns the model using the reservation status data extracted by the extraction unit 112 as learning data.
  • learning data can be used as a learning method.
  • Learning a model can refer to adjusting the parameters (eg, weights and biases) that make up the model.
  • the model is configured to take as input applicant's characteristic information and candidate date information and output a score for each of the candidate dates.
  • V represents candidate date information
  • c represents characteristic information
  • e represents a score for each of the candidate dates.
  • Only characteristics for which conditions are set in the weight information may be used as the characteristic information c.
  • the characteristic information c may relate to multiple characteristics including age, gender, and voice tone, or may relate to age alone.
  • the score represents the applicant's assumed degree of satisfaction with the construction of the optical fiber line on the candidate date.
  • Let f0 be the distribution of construction dates (specifically, the number of days from the application date to the construction date) obtained when it is assumed that construction reservations can be freely executed for reservation status data.
  • the distribution f0 can be obtained based on the information indicating the application timing included in the reservation status data.
  • the learning unit 113 learns the model so that the higher the frequency (the number of workers assigned to the work) in the distribution f0 , the higher the score.
  • the reservation status data includes property information of the person making the reservation, information indicating the requested service, information indicating the time of application, candidate date information, and a score for the construction date. and including.
  • the number of days from the application date to the construction date calculated from the reservation person's characteristic information, the candidate date information, the information indicating the application time, and the score are used as learning data.
  • the person in charge of reception asks the customer a questionnaire about the customer's degree of satisfaction with the reservation result, and registers the content of the reservation including the customer's degree of satisfaction in the reservation status storage unit 13 .
  • the learning unit 113 aggregates customer satisfaction levels for each period (number of days) from the application date to the construction date, and calculates a score from the aggregated result. If there is no information similar to the score, the learning unit 113 calculates the distribution of construction dates in the same manner as described above, and calculates the score from the frequency in the distribution.
  • the model generation unit 11 generates a model each time weight information is input, and stores the generated model in the model storage unit 14 in association with the weight information.
  • FIG. 4 schematically shows a configuration example of the information presentation unit 12.
  • the information presentation unit 12 includes an acquisition unit 121, a selection unit 122, a prediction unit 123, and an output unit .
  • the acquisition unit 121 acquires applicant information including the applicant's characteristic information and information indicating the application service.
  • Applicant information is entered, for example, by a receptionist.
  • the receptionist evaluates the applicant's tone of voice from the applicant's voice, and inputs the voice tone evaluation result (normal/displeased) along with the applicant's age and gender.
  • the applicant's tone of voice may be evaluated by analyzing the applicant's voice with voice analysis software.
  • the selection unit 122 selects a model from a plurality of models stored in the model storage unit 14. Model selection may be performed based on the applicant information acquired by the acquisition unit 121 .
  • the selection unit 122 selects, from among a plurality of models, a model associated with weight information that matches the characteristic information included in the applicant information. For example, when the characteristic information of the applicant is information that the age is 25 years old, the gender is male, and the tone of voice is sullen, the selection unit 122 selects a range of 20 to 29 years old for the characteristic "age”, Select a model that is associated with weight information that includes a condition specifying male for the characteristic "sex" and moody for the characteristic "voice".
  • the selection unit 122 selects the model associated with the weight information that has a large number of matching characteristics with the characteristic information. good.
  • a second model associated with weight information that includes a specified condition.
  • the selection unit 122 selects the first model.
  • Model selection may be performed based on the instructions of the receptionist.
  • a user interface screen for model selection is displayed on the display device of the computer terminal used by the person in charge of reception.
  • a user interface screen presents the weight information associated with each model.
  • the receptionist refers to the presented weight information to specify the appropriate model for the applicant. For example, when the receptionist recognizes that the applicant is in a bad mood, he selects a model associated with weight information in which the value "displeased" is set for the characteristic "voice tone.”
  • the prediction unit 123 uses the model selected by the selection unit 122 to predict the desired construction date based on the applicant information acquired by the acquisition unit 121 . Specifically, the prediction unit 123 inputs the characteristic information included in the applicant information and the candidate date information indicating the candidate date for the optical line installation to the model, and obtains the score for each candidate date from the model. The prediction unit 123 may identify the candidate date with the highest score as the first candidate.
  • the output unit 124 outputs the prediction result of the desired construction date.
  • the prediction result includes, for example, information indicating the candidate date identified as the first candidate and scores for each of the other candidate dates.
  • FIG. 5 schematically shows a hardware configuration example of the information processing device 10.
  • the information processing apparatus 10 includes a processor 51, a RAM (Random Access Memory) 52, a program memory 53, a storage device 54, and an input/output interface 55 as hardware elements.
  • Processor 51 is communicatively connected to RAM 52 , program memory 53 , storage device 54 and input/output interface 55 .
  • the processor 51 includes general-purpose circuits such as a CPU (Central Processing Unit).
  • RAM 52 includes volatile memory such as SDRAM.
  • RAM 52 is used by processor 51 as a working memory.
  • Program memory 53 stores programs executed by processor 51, including a model generation program and an information presentation program.
  • the program includes computer-executable instructions.
  • a ROM for example, is used as the program memory 53 .
  • a partial area of the storage device 54 may be used as the program memory 53 .
  • the processor 51 expands the program stored in the program memory 53 to the RAM 52, interprets and executes the program.
  • the model generation program causes the processor 51 to perform a series of processes described with respect to the model generation unit 11 .
  • the information presentation program when executed by the processor 51 , causes the processor 51 to perform a series of processes described with respect to the information presentation section 12 .
  • the program may be provided to the information processing device 10 while being stored in a computer-readable recording medium.
  • the information processing apparatus 10 has a drive for reading data from the recording medium, and acquires the program from the recording medium.
  • Examples of recording media include magnetic disks, optical disks (CD-ROM, CD-R, DVD-ROM, DVD-R, etc.), magneto-optical disks (MO, etc.), and semiconductor memories.
  • the program may be distributed through a network. Specifically, the program may be stored in a server on the network, and the information processing apparatus 10 may download the program from the server.
  • the storage device 54 includes non-volatile memory such as HDD (Hard Disk Drive) or SSD (Solid State Drive).
  • the storage device 54 stores data such as reservation status data and models.
  • the input/output interface 55 includes a communication module for communicating with an external device and a plurality of terminals for connecting peripheral devices.
  • Communication modules include wired modules and/or wireless modules. Examples of peripherals include displays, keyboards, and mice.
  • the processor 51 acquires data such as weight information and applicant information through the input/output interface 55 .
  • the processor 51 outputs the prediction result of the desired construction date via the input/output interface 55 .
  • FIG. 6 schematically shows an example of a process in which the model generation unit 11 generates a model.
  • the acquisition unit 111 acquires weight information indicating a weight for at least one characteristic of the reservation person. For example, a user who manages the information processing apparatus 10 inputs weight information including a condition specifying a displeased voice tone.
  • step S12 the extraction unit 112 extracts reservation status data corresponding to the weight information acquired by the acquisition unit 111 from the reservation status data stored in the reservation status storage unit 13. Specifically, the extraction unit 112 acquires reservation status data including characteristic information that matches the conditions regarding each characteristic included in the weight information from the reservation status storage unit 13 . For example, the extraction unit 112 reads, from the reservation status storage unit 13, reservation status data including characteristic information in which the value of the characteristic “voice tone” is “displeased”.
  • the learning unit 113 uses the reservation status data extracted by the extraction unit 112 as learning data to perform supervised learning of the model.
  • the model is configured to receive as input characteristic information of applicants and candidate date information indicating candidate dates for optical line installation, and to output scores for each of the candidate dates.
  • the learning unit 113 learns the model so that the higher the frequency in the construction date distribution obtained from the reservation status data extracted by the extraction unit 112, the higher the score.
  • step S14 the learning unit 113 outputs a trained model.
  • the learning unit 113 stores the learned model in the model storage unit 14 in association with the weight information.
  • FIG. 7 schematically shows an example of processing in which the information presentation unit 12 presents information.
  • the acquisition unit 121 acquires applicant information including characteristic information of the applicant.
  • step S22 If model selection is performed based on the applicant information (step S22; Yes), the flow proceeds to step S23; if model selection is performed based on the instructions of the receptionist (step S22; , the flow proceeds to step S24.
  • the selection unit 122 displays a user interface screen for selecting a model on the display device.
  • a user interface screen lists selectable models. Weight information associated with the model is displayed as information representing the feature of each model.
  • the user interface screen includes a button for automatically selecting a model. If the receptionist knows which model to use for the applicant, she selects one of the displayed models, otherwise she clicks a button. For example, when the receptionist determines that the applicant is displeased, he selects a model associated with weight information that includes a condition specifying displeased for the characteristic "voice tone.”
  • step S23 the selection unit 122 selects the model corresponding to the applicant information acquired in step S21 from among the models stored in the model storage unit 14. Specifically, the selection unit 122 searches the model storage unit 14 for weight information that matches the characteristic information included in the applicant information, and associates weight information that matches the characteristic information from the model storage unit 14 with the weight information that matches the characteristic information. Read the model that is
  • step S24 the selection unit 122 selects one of the models stored in the model storage unit 14 according to the instructions of the person in charge of reception. For example, the selection unit 122 reads the model selected by the receptionist from the model storage unit 14 .
  • step S25 the prediction unit 123 uses the model selected in step S23 or S24 to predict the applicant's desired construction date based on the applicant information.
  • the prediction unit 123 inputs the applicant's characteristic information and candidate date information included in the applicant information into the model, and obtains a score for each candidate date from the model.
  • the prediction unit 123 may identify the candidate date with the highest score as the first candidate.
  • the prediction unit 123 may further identify the candidate date with the second highest score as the second candidate, and may further identify the candidate date with the third highest score as the third candidate.
  • the output unit 124 outputs the prediction result of the desired construction date.
  • the output unit 124 displays the prediction result of the desired construction date on the display device.
  • the prediction result of the desired construction date may include information indicating the candidate date identified as the first candidate and scores for each of the other candidate dates.
  • the desired construction date prediction results may include a score for each of all candidate dates.
  • the prediction result of the desired construction date may include information indicating the candidate date specified as the first candidate, the candidate date specified as the second candidate, and the candidate date specified as the third candidate.
  • the information processing device 10 acquires weight information indicating a weight for at least one characteristic of a reservation person, extracts reservation status data corresponding to the weight information from reservation status data related to a plurality of reservation persons, and extracts the extracted reservation status data. Using the reservation status data as learning data, train a model to predict the applicant's desired construction date.
  • the model is configured to receive as input characteristic information of applicants and candidate date information indicating candidate dates for optical line installation, and to output scores for each of the candidate dates. The score represents the applicant's assumed degree of satisfaction with the construction of the optical fiber line on the candidate date.
  • the information processing apparatus 10 sets the model so that the more frequent the day, the higher the score in the distribution of work days obtained when it is assumed that the reservation for the optical line work can be freely executed for the extracted reservation status data. to learn.
  • a model corresponds to a utility function, which is a function that converts the degree of satisfaction with a product or service into a numerical value. The model generated in this way makes it possible to efficiently propose a construction date to applicants who apply for reservations for optical line construction.
  • At least one characteristic may include timbre. This makes it possible to more appropriately propose a construction date in a telephone-based reservation system.
  • the information processing apparatus 10 associates a plurality of models with weight information and stores them in the model storage unit 14, acquires characteristic information indicating characteristics of the applicant, selects a model from among the plurality of models, and stores the characteristics of the applicant. Information and candidate date information are input to the selected model, based on the score output from the selected model, the applicant's desired construction date is predicted, and the prediction result is output.
  • the information processing device 10 may select a model associated with weight information corresponding to the applicant's characteristic information from among a plurality of models.
  • the person in charge of reception refers to the prediction result output from the information processing device 10 in order to propose the construction date to the applicant. According to the information processing device 10, it is possible to make the proposal of the construction date by the person in charge of reception more efficient.
  • the degree of satisfaction of the applicant can be estimated from the score on the construction date agreed upon by the applicant. For example, it can be inferred that the applicant's satisfaction level is low if the score on the construction date agreed upon by the applicant is lower than a threshold. In that case, the receptionist can take some action, such as suggesting additional services or responding more politely.
  • the score (expected satisfaction) for each candidate date is used as a criterion for deciding whether additional services should be offered or whether a more polite response is required when the applicant's wishes cannot be realized. be able to.
  • price can be used as an effective indicator for inferring the utility function. It is possible to infer the utility function for each customer by presenting different prices to each customer on each day and obtaining feedback of customer selection.
  • this embodiment it is not necessary to be able to use effective indicators to infer utility functions such as prices. Therefore, this embodiment can also be applied to reservations for services in the communication industry as described above. Even for an applicant who does not have a definite desired construction date, the possibility of being able to present a highly satisfying construction date increases, and it is possible to shorten the time until the proposal is made.
  • the present invention is not limited to the above-described embodiments, and can be variously modified in the implementation stage without departing from the gist of the present invention. Further, each embodiment may be implemented in combination as appropriate, in which case the combined effect can be obtained. Furthermore, various inventions are included in the above embodiments, and various inventions can be extracted by combinations selected from the disclosed plurality of components. For example, even if some components are deleted from all the components shown in the embodiment, if the problem can be solved and effects can be obtained, the configuration in which these components are deleted can be extracted as an invention.

Landscapes

  • Business, Economics & Management (AREA)
  • Engineering & Computer Science (AREA)
  • Tourism & Hospitality (AREA)
  • Operations Research (AREA)
  • Economics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Human Resources & Organizations (AREA)
  • Marketing (AREA)
  • Development Economics (AREA)
  • Quality & Reliability (AREA)
  • Strategic Management (AREA)
  • Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

This information processing device is provided with: a first acquisition unit which acquires weight information indicating weights for at least one feature of a reservation holder; an extraction unit which extracts reservation status data corresponding to the aforementioned weight information from reservation status data relating to multiple reservation holders, wherein the reservation status data relating to each of the reservation holders indicates the content of completed reservations of the reservation holder and includes feature information indicating features of the reservation holder; and a learning unit which uses the extracted reservation status data as training data, and which trains a model that inputs feature information indicating features of an applicant and candidate day information indicating candidate days for performing a service, and that, for each of the candidate days, outputs a score representing the estimated satisfaction of the applicant in response to the service being carried out on that candidate day.

Description

情報処理装置、情報処理方法、及びプログラムInformation processing device, information processing method, and program
　本発明は、情報処理装置、情報処理方法、及びプログラムに関する。 The present invention relates to an information processing device, an information processing method, and a program.
　顧客の所有する端末装置からレストランや宿泊施設などのサービスに対する予約を申し込む予約システムが知られている。そのような予約システムでは、サービス提供側は、予約申込時に、顧客の年齢、性別、住所などの属性を示す属性情報を、顧客を特定する情報として得ている。 Reservation systems are known for applying for reservations for services such as restaurants and lodging facilities from terminal devices owned by customers. In such a reservation system, the service provider obtains attribute information indicating the customer's age, sex, address, and other attributes as information identifying the customer when applying for a reservation.
　予約申込の簡便性を得るために、顧客を特定する情報として顧客の電話番号を利用することがある。その場合には、サービス提供側は顧客の属性情報を得ることができない。特許文献１は、顧客を特定する情報として顧客の電話番号を利用する予約システムにおいて、顧客の属性を把握することを可能にする技術を開示している。特許文献１に開示される技術では、顧客が自身の属性情報を提供することに対する見返りとして、顧客に対して特典を付与する。 　In order to facilitate reservation applications, the customer's telephone number may be used as information that identifies the customer. In that case, the service provider cannot obtain customer attribute information. Patent Literature 1 discloses a technology that makes it possible to grasp a customer's attribute in a reservation system that uses a customer's telephone number as information that identifies the customer. The technology disclosed in Patent Literature 1 gives a benefit to a customer in return for the customer's provision of his/her own attribute information.
　予約申込時に得られる顧客の属性情報を予約業務の品質の向上につなげるのは難しい。 　It is difficult to improve the quality of reservation work with the customer attribute information obtained at the time of reservation application.
日本国特開２００５－３４６４４１号公報Japanese Patent Application Laid-Open No. 2005-346441
　本発明は、予約業務の品質を向上させる技術を提供することを目的とする。 The purpose of the present invention is to provide technology for improving the quality of reservation operations.
　本発明の一態様に係る情報処理装置は、予約者の少なくとも１つの特性に対する重みを示す重み情報を取得する第１の取得部と、複数の予約者に関する予約状況データの中から、前記重み情報に対応する予約状況データを抽出する抽出部であって、前記複数の予約者の各々に関する予約状況データは、前記予約者について完了済みの予約の内容を示し、前記予約者の特性を示す特性情報を含む、抽出部と、前記抽出された予約状況データを学習データとして使用して、申込者の特性を示す特性情報とサービス実施の候補日を示す候補日情報とを入力とし、前記候補日のそれぞれに対する、サービスが候補日に実施されることに対する前記申込者の想定満足度を表すスコアを出力するモデルを学習する学習部と、を備える。 An information processing apparatus according to an aspect of the present invention includes: a first acquisition unit that acquires weight information indicating a weight for at least one characteristic of a reservation person; wherein the reservation status data relating to each of the plurality of reserving persons indicates the details of reservations completed for the reserving person, and characteristic information indicating characteristics of the reserving person. and an extraction unit, using the extracted reservation status data as learning data, inputting characteristic information indicating characteristics of the applicant and candidate date information indicating a candidate date for service implementation, and inputting the candidate date for each, a learning unit that trains a model that outputs a score representing the applicant's expected satisfaction with the service being performed on the candidate date.
　本発明によれば、予約業務の品質を向上させる技術を提供することができる。 According to the present invention, it is possible to provide technology for improving the quality of reservation operations.
図１は、本発明の実施形態に係る情報処理装置を示すブロック図である。FIG. 1 is a block diagram showing an information processing apparatus according to an embodiment of the invention. 図２は、図１に示したモデル生成部の構成を示すブロック図である。FIG. 2 is a block diagram showing the configuration of the model generating unit shown in FIG. 1; 図３は、図１に示したモデル生成部により生成されるモデルから出力されるスコアを説明する図である。FIG. 3 is a diagram for explaining scores output from a model generated by the model generation unit shown in FIG. 図４は、図１に示した情報提示部の構成を示すブロック図である。FIG. 4 is a block diagram showing the configuration of the information presentation unit shown in FIG. 図５は、図１に示した情報処理装置のハードウェア構成を示すブロック図である。FIG. 5 is a block diagram showing the hardware configuration of the information processing apparatus shown in FIG. 1. As shown in FIG. 図６は、本発明の実施形態に係るモデル生成の手順を示すフローチャートである。FIG. 6 is a flow chart showing the procedure of model generation according to the embodiment of the present invention. 図７は、本発明の実施形態に係る情報提示の手順を示すフローチャートである。FIG. 7 is a flow chart showing the information presentation procedure according to the embodiment of the present invention.
　以下、図面を参照して本発明の実施形態を説明する。本発明の実施形態は、顧客からサービスに対する予約の申し込みを受け付けて、顧客の希望に沿った予約を行う予約システムに適用可能な情報処理装置に関する。以降では、予約の申し込みを行う顧客を申込者と称する。また、予約が完了した顧客を予約者と称する。 Hereinafter, embodiments of the present invention will be described with reference to the drawings. An embodiment of the present invention relates to an information processing apparatus applicable to a reservation system that accepts a reservation application for a service from a customer and makes a reservation according to the customer's request. Hereinafter, a customer who applies for a reservation is referred to as an applicant. A customer who has completed a reservation is referred to as a reservation person.
　ここでは、光回線工事を予約する予約システムを例に挙げて実施形態の説明を行う。自宅で光回線インターネットサービスを利用することを望む申込者は、インターネットサービスプロバイダに対し、光回線インターネットサービスの利用を申し込む。光回線インターネットサービスを利用するためには、申込者の住宅に光ファイバを引き込むための光回線工事が必要になることがある。光回線工事には、申込者又はその関係者の立ち会いが必要とされる。インターネットサービスプロバイダの受付担当者は、電話にて申込者との間で光回線工事の日程を調整する。 Here, the embodiment will be explained using a reservation system for making reservations for optical line construction as an example. An applicant who wants to use the optical line internet service at home applies for the use of the optical line internet service to an internet service provider. In order to use the optical line Internet service, it may be necessary to install an optical line to bring an optical fiber into the applicant's residence. The presence of the applicant or a person concerned is required for the optical line construction. The receptionist of the Internet service provider adjusts the schedule of the optical line construction with the applicant by telephone.
　図１は、本発明の実施形態に係る情報処理装置１０を概略的に示している。図１に示すように、情報処理装置１０は、モデル生成部１１、情報提示部１２、予約状況記憶部１３、及びモデル記憶部１４を備える。 FIG. 1 schematically shows an information processing device 10 according to an embodiment of the invention. As shown in FIG. 1, the information processing device 10 includes a model generation unit 11, an information presentation unit 12, a reservation status storage unit 13, and a model storage unit .
　予約状況記憶部１３は複数の予約者のそれぞれに関する予約状況データを格納する。予約状況データは完了済みの予約（過去の予約）の内容を示すデータである。例えば、予約状況データは予約者の特性情報及び申込情報を含む。特性情報は顧客（例えば申込者又は予約者）の１以上の特性を示す情報である。特性の例は、年齢、性別、声色（機嫌）を含む。声色は例えば普通及び不機嫌（怒り）という２つの値で表される。なお、声色は３つ以上の値で表されてもよい。声色は、予約時に予約者と通話した受付担当者により評価されてもよく、予約時の予約者の通話音声を音声分析ソフトウェアで分析することにより評価されてもよい。申込情報は、例えば、申込サービスを示す情報と、申込時期を示す情報と、サービス実施の候補日を示す候補日情報と、を含む。ここで参照する例では、申込サービスは光回線工事である。申込時期は、例えば、光回線工事の予約を行った申込日と、光回線工事が行われることになる工事日と、を含む。申込時期を示す情報に基づいて、申し込みから光回線工事の実施までの日数を算出することが可能である。候補日情報は、光回線工事の候補日を示す。光回線工事の候補日は、空き工事日、すなわち、インターネットサービスプロバイダが光回線工事を実施することが可能な日を示す。ここでは、候補日は、申込日からの日数で表される。予約が情報処理装置１０を使用して行われた場合には、申込情報は、工事日に対するスコアを含む。スコアについては後述する。 The reservation status storage unit 13 stores reservation status data for each of a plurality of subscribers. The reservation status data is data indicating the details of completed reservations (past reservations). For example, the reservation status data includes characteristic information and application information of the reservation person. Characteristic information is information that indicates one or more characteristics of a customer (eg, applicant or reservation person). Examples of characteristics include age, gender, tone of voice (mood). Voice tone is represented by two values, for example normal and displeased (angry). Note that the voice tone may be represented by three or more values. The tone of voice may be evaluated by the receptionist who spoke with the person who made the reservation at the time of the reservation, or by analyzing the voice of the person who made the reservation at the time of the reservation with voice analysis software. The application information includes, for example, information indicating an application service, information indicating an application time, and candidate date information indicating a candidate date for service implementation. In the example referred to here, the application service is optical line construction. The application time includes, for example, the application date when the reservation for the optical line construction was made and the construction date when the optical line construction is to be performed. Based on the information indicating the application period, it is possible to calculate the number of days from the application to the implementation of the optical line construction. The candidate date information indicates the candidate date for optical line installation. The candidate date for optical line installation indicates a free installation date, that is, a date on which the Internet service provider can perform the optical line installation. Here, the candidate date is represented by the number of days from the application date. When the reservation is made using the information processing device 10, the application information includes the score for the construction date. Scores will be discussed later.
　モデル生成部１１は、予約状況記憶部１３に格納されている予約状況データを使用して、申込者が光回線工事の実施を希望する日である希望工事日を予測するために使用する機械学習モデルを生成して、生成した機械学習モデルをモデル記憶部１４に格納する。モデル記憶部１４は、モデル生成部１１により生成される複数のモデルを格納する。 The model generation unit 11 uses the reservation status data stored in the reservation status storage unit 13 to predict the desired construction date, which is the date on which the applicant wishes to perform the optical line construction. A model is generated and the generated machine learning model is stored in the model storage unit 14 . The model storage unit 14 stores multiple models generated by the model generation unit 11 .
　情報提示部１２は、モデル記憶部１４に格納されている複数のモデルのうちのいずれかのモデルを使用して、申込者情報に基づいて申込者の希望工事日を予測し、予測結果を出力する。申込者情報は、申込者の特性情報と、申込サービスを示す情報と、を含む。受付担当者は、情報提示部１２により出力される予測結果を参照して工事日を申込者に提案することができる。予約（日程調整）が完了した後に、受付担当者は予約の内容を予約状況記憶部１３に登録する。 The information presentation unit 12 uses one of a plurality of models stored in the model storage unit 14 to predict the applicant's desired construction date based on the applicant information, and outputs the prediction result. do. Applicant information includes applicant's characteristic information and information indicating an applied service. The person in charge of reception can refer to the prediction result output by the information presentation unit 12 and propose the construction date to the applicant. After completing the reservation (schedule adjustment), the person in charge of reception registers the contents of the reservation in the reservation status storage unit 13 .
　図１に示す例では、モデル生成部１１及び情報提示部１２は１つの装置に存在する。代替として、モデル生成部１１及び情報提示部１２は別々の装置に存在してもよい。 In the example shown in FIG. 1, the model generation unit 11 and the information presentation unit 12 exist in one device. Alternatively, the model generator 11 and the information presenter 12 may exist in separate devices.
　図２は、モデル生成部１１の構成例を概略的に示している。図２に示すように、モデル生成部１１は、取得部１１１、抽出部１１２、及び学習部１１３を備える。 FIG. 2 schematically shows a configuration example of the model generation unit 11. As shown in FIG. As shown in FIG. 2 , the model generation unit 11 includes an acquisition unit 111 , an extraction unit 112 and a learning unit 113 .
　取得部１１１は、予約者の少なくとも１つの特性に対する重みを示す重み情報を取得する。重み情報は情報処理装置１０を管理する利用者により入力され得る。重み情報は、各特性に関する値又は範囲を指定する条件を含んでよい。重み情報は申込サービスを指定する情報をさらに含んでよい。 The acquisition unit 111 acquires weight information indicating a weight for at least one characteristic of the reservation person. The weight information can be input by a user who manages the information processing device 10 . Weight information may include conditions that specify values or ranges for each property. The weight information may further include information specifying the subscription service.
　抽出部１１２は、予約状況記憶部１３に格納されている予約状況データから、取得部１１１により取得される重み情報に対応する予約状況データを抽出する。例えば、抽出部１１２は、予約状況記憶部１３から、重み情報に含まれる条件に合致する特性情報を含む予約状況データを取得する。例えば、重み情報が声色について不機嫌を指定する条件を含む場合、抽出部１１２は、予約状況記憶部１３から、特性「声色」の値が「不機嫌」である特性情報を含む予約状況データを取得する。例えば、重み情報が、年齢について２０～２９歳の範囲、性別について男、声色について不機嫌を指定する条件を含む場合、抽出部１１２は、予約状況記憶部１３から、特性「年齢」の値が２０～２９のいずれかであり、特性「性別」の値が「男」であり、特性「声色」の値が「不機嫌」である特性情報を含む予約状況データを取得する。上述したように、予約状況データは特性情報に加えて申込情報を含む。 The extraction unit 112 extracts reservation status data corresponding to the weight information acquired by the acquisition unit 111 from the reservation status data stored in the reservation status storage unit 13 . For example, the extraction unit 112 acquires reservation status data including characteristic information that matches the conditions included in the weight information from the reservation status storage unit 13 . For example, if the weight information includes a condition specifying a displeased voice tone, the extraction unit 112 acquires reservation status data including characteristic information in which the value of the characteristic "voice tone" is "displeased" from the reservation status storage unit 13. . For example, if the weight information includes a condition that designates the age range of 20 to 29, the gender as male, and the tone of voice as displeased, the extraction unit 112 extracts from the reservation status storage unit 13 the value of the characteristic “age” of 20. 29, the value of the property “sex” is “male”, and the value of the property “voice tone” is “sullen”. As described above, reservation status data includes offer information in addition to property information.
　学習部１１３は、抽出部１１２により抽出される予約状況データを学習データとして使用してモデルを学習する。学習手法としては、機械学習又はディープラーニングを使用することができる。モデルを学習するとは、モデルを構成するパラメータ（例えば重み及びバイアス）を調整することを指し得る。 The learning unit 113 learns the model using the reservation status data extracted by the extraction unit 112 as learning data. As a learning method, machine learning or deep learning can be used. Learning a model can refer to adjusting the parameters (eg, weights and biases) that make up the model.
　モデルは、申込者の特性情報と候補日情報とを入力とし、候補日のそれぞれに対するスコアを出力するように構成される。モデルは、例えば、下記のように表すことができる。
　　ｅ＝ｆ（Ｖ，ｃ）
　ここで、Ｖは候補日情報を表し、ｃは特性情報を表し、ｅは候補日のそれぞれに対するスコアを表す。特性情報ｃとして、重み情報において条件が設定される特性のみが使用されてもよい。例えば重み情報が年齢に対する条件を含む場合、特性情報ｃは、年齢、性別、及び声色を含む複数の特性に関するものであってもよく、年齢のみに関するものであってもよい。 The model is configured to take as input applicant's characteristic information and candidate date information and output a score for each of the candidate dates. The model can be represented, for example, as follows.
e=f(V,c)
Here, V represents candidate date information, c represents characteristic information, and e represents a score for each of the candidate dates. Only characteristics for which conditions are set in the weight information may be used as the characteristic information c. For example, if the weight information includes a condition for age, the characteristic information c may relate to multiple characteristics including age, gender, and voice tone, or may relate to age alone.
　スコアは、光回線工事が候補日に実施されることに対する申込者の想定満足度を表す。予約状況データに対して自由に工事予約を実行できると仮定した場合に得られる工事日（具体的には申込日から工事日までの日数）の分布をｆ０とする。分布ｆ０は予約状況データに含まれる申込時期を示す情報に基づいて求めることができる。学習部１１３は、図３に示すように、分布ｆ０において頻度（工事割り当て人数）が多い日ほどスコアが高くなるように、モデルを学習する。 The score represents the applicant's assumed degree of satisfaction with the construction of the optical fiber line on the candidate date. Let f0 be the distribution of construction dates (specifically, the number of days from the application date to the construction date) obtained when it is assumed that construction reservations can be freely executed for reservation status data. The distribution f0 can be obtained based on the information indicating the application timing included in the reservation status data. As shown in FIG. 3, the learning unit 113 learns the model so that the higher the frequency (the number of workers assigned to the work) in the distribution f0 , the higher the score.
　予約が情報処理装置１０を使用して行われた場合、予約状況データは、予約者の特性情報と、申込サービスを示す情報と、申込時期を示す情報と、候補日情報と、工事日に対するスコアと、を含む。予約者の特性情報、候補日情報、申込時期を示す情報から算出される申込日から工事日までの日数、及びスコアが学習データとして使用される。 When a reservation is made using the information processing device 10, the reservation status data includes property information of the person making the reservation, information indicating the requested service, information indicating the time of application, candidate date information, and a score for the construction date. and including. The number of days from the application date to the construction date calculated from the reservation person's characteristic information, the candidate date information, the information indicating the application time, and the score are used as learning data.
　最初のモデルを作成する場合などのように予約状況データにスコアが含まれない場合には、スコアに類する情報があれば、その情報を用いてスコアを決定する。例えば、受付担当者は、顧客に予約結果に対する満足度をアンケートし、顧客の満足度を含む予約の内容を予約状況記憶部１３に登録する。学習部１１３は、申込日から工事日までの期間（日数）別で、顧客の満足度を集計し、集計結果からスコアを算出する。スコアに類する情報がない場合、学習部１１３は、上述したものと同様にして、工事日の分布を算出し、分布における頻度からスコアを算出する。 When the reservation status data does not contain the score, such as when creating the first model, if there is information similar to the score, that information is used to determine the score. For example, the person in charge of reception asks the customer a questionnaire about the customer's degree of satisfaction with the reservation result, and registers the content of the reservation including the customer's degree of satisfaction in the reservation status storage unit 13 . The learning unit 113 aggregates customer satisfaction levels for each period (number of days) from the application date to the construction date, and calculates a score from the aggregated result. If there is no information similar to the score, the learning unit 113 calculates the distribution of construction dates in the same manner as described above, and calculates the score from the frequency in the distribution.
　モデル生成部１１は、重み情報が入力されるたびにモデルを生成し、生成したモデルを重み情報に関連付けてモデル記憶部１４に格納する。 The model generation unit 11 generates a model each time weight information is input, and stores the generated model in the model storage unit 14 in association with the weight information.
　図４は、情報提示部１２の構成例を概略的に示している。図４に示すように、情報提示部１２は、取得部１２１、選択部１２２、予測部１２３、及び出力部１２４を備える。 FIG. 4 schematically shows a configuration example of the information presentation unit 12. As shown in FIG. As shown in FIG. 4, the information presentation unit 12 includes an acquisition unit 121, a selection unit 122, a prediction unit 123, and an output unit .
　取得部１２１は、申込者の特性情報と申込サービスを示す情報とを含む申込者情報を取得する。申込者情報は例えば受付担当者により入力される。例えば、受付担当者は、申込者の音声から申込者の声色を評価し、申込者の年齢及び性別とともに声色の評価結果（普通／不機嫌）を入力する。なお、申込者の声色は、申込者の音声を音声分析ソフトウェアで分析することにより評価されてもよい。 The acquisition unit 121 acquires applicant information including the applicant's characteristic information and information indicating the application service. Applicant information is entered, for example, by a receptionist. For example, the receptionist evaluates the applicant's tone of voice from the applicant's voice, and inputs the voice tone evaluation result (normal/displeased) along with the applicant's age and gender. Note that the applicant's tone of voice may be evaluated by analyzing the applicant's voice with voice analysis software.
　選択部１２２は、モデル記憶部１４に格納されている複数のモデルの中からモデルを選択する。モデル選択は、取得部１２１により取得される申込者情報に基づいて実行されてよい。選択部１２２は、複数のモデルの中から、申込者情報に含まれる特性情報に合致する重み情報に関連付けられているモデルを選択する。例えば、申込者の特性情報が、年齢が２５歳であり、性別が男であり、声色が不機嫌であるという情報である場合、選択部１２２は、特性「年齢」について２０～２９歳の範囲、特性「性別」について男、特性「声色」について不機嫌を指定する条件を含む重み情報に関連付けられているモデルを選択する。 The selection unit 122 selects a model from a plurality of models stored in the model storage unit 14. Model selection may be performed based on the applicant information acquired by the acquisition unit 121 . The selection unit 122 selects, from among a plurality of models, a model associated with weight information that matches the characteristic information included in the applicant information. For example, when the characteristic information of the applicant is information that the age is 25 years old, the gender is male, and the tone of voice is sullen, the selection unit 122 selects a range of 20 to 29 years old for the characteristic "age", Select a model that is associated with weight information that includes a condition specifying male for the characteristic "sex" and moody for the characteristic "voice".
　申込者情報に含まれる特性情報に合致する重み情報が複数ある場合には、選択部１２２は、特性情報との間で合致する特性の数が多い重み情報に関連付けられているモデルを選択してよい。特性「年齢」について２０～２９歳の範囲、特性「性別」について男、特性「声色」について不機嫌を指定する条件を含む重み情報に関連付けられている第１のモデルと特性「声色」について不機嫌を指定する条件を含む重み情報に関連付けられている第２のモデルとがあるとする。申込者の特性情報が、年齢が２５歳であり、性別が男であり、声色が不機嫌であるという情報である場合、第１のモデルについては、特性情報と重み情報との間で合致する特性の数は３であり、第２のモデルについては、特性情報と重み情報との間で合致する特性の数は１である。この場合、選択部１２２は第１のモデルを選択する。 If there are multiple pieces of weight information that match the characteristic information included in the applicant information, the selection unit 122 selects the model associated with the weight information that has a large number of matching characteristics with the characteristic information. good. A first model associated with weight information that includes conditions specifying a range of 20-29 years for the characteristic "Age", male for the characteristic "Gender", and moody for the characteristic "Voice tone" and moody for the characteristic "Voice tone". Suppose there is a second model associated with weight information that includes a specified condition. If the characteristic information of the applicant is information that the age is 25 years old, the gender is male, and the voice tone is sullen, for the first model, the characteristics that match between the characteristic information and the weight information is three, and for the second model, the number of matching features between the feature information and the weight information is one. In this case, the selection unit 122 selects the first model.
　モデル選択は受付担当者の指示に基づいて実行されてもよい。受付担当者が使用するコンピュータ端末の表示装置にモデル選択のためのユーザインタフェース画面が表示される。ユーザインタフェース画面には、各モデルに関連付けられる重み情報が提示される。受付担当者は、提示される重み情報を参照して、申込者に適したモデルを指定する。例えば、受付担当者は、申込者が不機嫌であると認識した場合には、特性「声色」について値「不機嫌」が設定される重み情報に関連付けられているモデルを選択する。 　Model selection may be performed based on the instructions of the receptionist. A user interface screen for model selection is displayed on the display device of the computer terminal used by the person in charge of reception. A user interface screen presents the weight information associated with each model. The receptionist refers to the presented weight information to specify the appropriate model for the applicant. For example, when the receptionist recognizes that the applicant is in a bad mood, he selects a model associated with weight information in which the value "displeased" is set for the characteristic "voice tone."
　予測部１２３は、選択部１２２により選択されるモデルを使用して、取得部１２１により取得される申込者情報に基づいて、希望工事日を予測する。具体的には、予測部１２３は、申込者情報に含まれる特性情報と光回線工事の候補日を示す候補日情報とをモデルに入力し、モデルから、候補日のそれぞれに対するスコアを得る。予測部１２３は、スコアが最も高い候補日を第１候補として特定してよい。 The prediction unit 123 uses the model selected by the selection unit 122 to predict the desired construction date based on the applicant information acquired by the acquisition unit 121 . Specifically, the prediction unit 123 inputs the characteristic information included in the applicant information and the candidate date information indicating the candidate date for the optical line installation to the model, and obtains the score for each candidate date from the model. The prediction unit 123 may identify the candidate date with the highest score as the first candidate.
　出力部１２４は、希望工事日の予測結果を出力する。予測結果は、例えば、第１候補として特定された候補日と、他の候補日のそれぞれに対するスコアと、を示す情報を含む。 The output unit 124 outputs the prediction result of the desired construction date. The prediction result includes, for example, information indicating the candidate date identified as the first candidate and scores for each of the other candidate dates.
　図５は、情報処理装置１０のハードウェア構成例を概略的に示している。図５に示すように、情報処理装置１０は、ハードウェア要素として、プロセッサ５１、ＲＡＭ（Random Access Memory）５２、プログラムメモリ５３、ストレージデバイス５４、及び入出力インタフェース５５を備える。プロセッサ５１は、ＲＡＭ５２、プログラムメモリ５３、ストレージデバイス５４、及び入出力インタフェース５５と通信可能に接続される。 FIG. 5 schematically shows a hardware configuration example of the information processing device 10. As shown in FIG. As shown in FIG. 5, the information processing apparatus 10 includes a processor 51, a RAM (Random Access Memory) 52, a program memory 53, a storage device 54, and an input/output interface 55 as hardware elements. Processor 51 is communicatively connected to RAM 52 , program memory 53 , storage device 54 and input/output interface 55 .
　プロセッサ５１は、ＣＰＵ（Central Processing Unit）などの汎用回路を含む。ＲＡＭ５２はＳＤＲＡＭなどの揮発性メモリを含む。ＲＡＭ５２はワーキングメモリとしてプロセッサ５１により使用される。プログラムメモリ５３は、モデル生成プログラムや情報提示プログラムを含む、プロセッサ５１により実行されるプログラムを記憶する。プログラムはコンピュータ実行可能命令を含む。プログラムメモリ５３として例えばＲＯＭが使用される。ストレージデバイス５４の一部領域がプログラムメモリ５３として使用されてもよい。 The processor 51 includes general-purpose circuits such as a CPU (Central Processing Unit). RAM 52 includes volatile memory such as SDRAM. RAM 52 is used by processor 51 as a working memory. Program memory 53 stores programs executed by processor 51, including a model generation program and an information presentation program. The program includes computer-executable instructions. A ROM, for example, is used as the program memory 53 . A partial area of the storage device 54 may be used as the program memory 53 .
　プロセッサ５１は、プログラムメモリ５３に記憶されたプログラムをＲＡＭ５２に展開し、プログラムを解釈及び実行する。モデル生成プログラムは、プロセッサ５１により実行されると、モデル生成部１１に関して説明される一連の処理をプロセッサ５１に行わせる。情報提示プログラムは、プロセッサ５１により実行されると、情報提示部１２に関して説明される一連の処理をプロセッサ５１に行わせる。 The processor 51 expands the program stored in the program memory 53 to the RAM 52, interprets and executes the program. When executed by the processor 51 , the model generation program causes the processor 51 to perform a series of processes described with respect to the model generation unit 11 . The information presentation program, when executed by the processor 51 , causes the processor 51 to perform a series of processes described with respect to the information presentation section 12 .
　プログラムは、コンピュータで読み取り可能な記録媒体に記憶された状態で情報処理装置１０に提供されてよい。この場合、情報処理装置１０は、記録媒体からデータを読み出すドライブを備え、記録媒体からプログラムを取得する。記録媒体の例は、磁気ディスク、光ディスク（ＣＤ－ＲＯＭ、ＣＤ－Ｒ、ＤＶＤ－ＲＯＭ、ＤＶＤ－Ｒなど）、光磁気ディスク（ＭＯなど）、及び半導体メモリを含む。また、プログラムはネットワークを通じて配布するようにしてもよい。具体的には、プログラムをネットワーク上のサーバに格納し、情報処理装置１０がサーバからプログラムをダウンロードするようにしてもよい。 The program may be provided to the information processing device 10 while being stored in a computer-readable recording medium. In this case, the information processing apparatus 10 has a drive for reading data from the recording medium, and acquires the program from the recording medium. Examples of recording media include magnetic disks, optical disks (CD-ROM, CD-R, DVD-ROM, DVD-R, etc.), magneto-optical disks (MO, etc.), and semiconductor memories. Also, the program may be distributed through a network. Specifically, the program may be stored in a server on the network, and the information processing apparatus 10 may download the program from the server.
　ストレージデバイス５４は、ＨＤＤ（Hard Disk Drive）又はＳＳＤ（Solid State Drive）などの不揮発性メモリを含む。ストレージデバイス５４は、予約状況データやモデルなどのデータを記憶する。 The storage device 54 includes non-volatile memory such as HDD (Hard Disk Drive) or SSD (Solid State Drive). The storage device 54 stores data such as reservation status data and models.
　入出力インタフェース５５は、外部装置と通信するための通信モジュールと、周辺機器を接続するための複数の端子と、を備える。通信モジュールは有線モジュール及び／又は無線モジュールを含む。周辺機器の例は、表示装置、キーボード、及びマウスを含む。プロセッサ５１は、入出力インタフェース５５を介して重み情報や申込者情報などのデータを取得する。プロセッサ５１は、入出力インタフェース５５を介して希望工事日の予測結果を出力する。 The input/output interface 55 includes a communication module for communicating with an external device and a plurality of terminals for connecting peripheral devices. Communication modules include wired modules and/or wireless modules. Examples of peripherals include displays, keyboards, and mice. The processor 51 acquires data such as weight information and applicant information through the input/output interface 55 . The processor 51 outputs the prediction result of the desired construction date via the input/output interface 55 .
　［動作］
　次に、情報処理装置１０の動作について説明する。 [motion]
Next, the operation of the information processing device 10 will be described.
　図６は、モデル生成部１１がモデルを生成する処理の一例を概略的に示している。図６のステップＳ１１において、取得部１１１は、予約者の少なくとも１つの特性に対する重みを示す重み情報を取得する。例えば、声色について不機嫌を指定する条件を含む重み情報が情報処理装置１０を管理する利用者により入力される。 FIG. 6 schematically shows an example of a process in which the model generation unit 11 generates a model. In step S11 of FIG. 6, the acquisition unit 111 acquires weight information indicating a weight for at least one characteristic of the reservation person. For example, a user who manages the information processing apparatus 10 inputs weight information including a condition specifying a displeased voice tone.
　ステップＳ１２において、抽出部１１２は、予約状況記憶部１３に格納されている予約状況データから、取得部１１１により取得された重み情報に対応する予約状況データを抽出する。具体的には、抽出部１１２は、予約状況記憶部１３から、重み情報に含まれる各特性に関する条件に合致する特性情報を含む予約状況データを取得する。例えば、抽出部１１２は、予約状況記憶部１３から、特性「声色」の値が「不機嫌」である特性情報を含む予約状況データを読み出す。 In step S12, the extraction unit 112 extracts reservation status data corresponding to the weight information acquired by the acquisition unit 111 from the reservation status data stored in the reservation status storage unit 13. Specifically, the extraction unit 112 acquires reservation status data including characteristic information that matches the conditions regarding each characteristic included in the weight information from the reservation status storage unit 13 . For example, the extraction unit 112 reads, from the reservation status storage unit 13, reservation status data including characteristic information in which the value of the characteristic “voice tone” is “displeased”.
　ステップＳ１３において、学習部１１３は、抽出部１１２により抽出された予約状況データを学習データとして使用してモデルの教師あり学習を行う。モデルは、申込者の特性情報と光回線工事の候補日を示す候補日情報とを入力とし、候補日のそれぞれに対するスコアを出力するように構成される。学習部１１３は、抽出部１１２により抽出された予約状況データから得られる工事日の分布において頻度が多い日ほどスコアが高くなるように、モデルを学習する。 In step S13, the learning unit 113 uses the reservation status data extracted by the extraction unit 112 as learning data to perform supervised learning of the model. The model is configured to receive as input characteristic information of applicants and candidate date information indicating candidate dates for optical line installation, and to output scores for each of the candidate dates. The learning unit 113 learns the model so that the higher the frequency in the construction date distribution obtained from the reservation status data extracted by the extraction unit 112, the higher the score.
　ステップＳ１４において、学習部１１３は、学習済みモデルを出力する。例えば、学習部１１３は、学習済みモデルを重み情報に関連付けてモデル記憶部１４に格納する。 In step S14, the learning unit 113 outputs a trained model. For example, the learning unit 113 stores the learned model in the model storage unit 14 in association with the weight information.
　図７は、情報提示部１２が情報を提示する処理の一例を概略的に示している。図７のステップＳ２１において、取得部１２１は申込者の特性情報を含む申込者情報を取得する。 FIG. 7 schematically shows an example of processing in which the information presentation unit 12 presents information. In step S21 of FIG. 7, the acquisition unit 121 acquires applicant information including characteristic information of the applicant.
　モデル選択が申込者情報に基づいて行われる場合（ステップＳ２２；Ｙｅｓ）には、フローはステップＳ２３に進み、モデル選択が受付担当者の指示に基づいて行われる場合（ステップＳ２２；Ｎｏ）には、フローはステップＳ２４に進む。例えば、選択部１２２は、モデルを選択するためのユーザインタフェース画面を表示装置に表示する。ユーザインタフェース画面は、選択可能なモデルを一覧表示する。各モデルの特徴を表す情報として、モデルに関連付けられている重み情報が表示される。ユーザインタフェース画面は、モデルを自動選択することを指示するボタンを含む。受付担当者は、申込者に対して使用すべきモデルを認識できている場合、表示されているモデルのうちの１つのモデルを選択し、そうでなければ、ボタンをクリックする。例えば、受付担当者は、申込者が不機嫌であると判断したときには、特性「声色」について不機嫌を指定する条件を含む重み情報に関連付けられているモデルを選択する。 If model selection is performed based on the applicant information (step S22; Yes), the flow proceeds to step S23; if model selection is performed based on the instructions of the receptionist (step S22; , the flow proceeds to step S24. For example, the selection unit 122 displays a user interface screen for selecting a model on the display device. A user interface screen lists selectable models. Weight information associated with the model is displayed as information representing the feature of each model. The user interface screen includes a button for automatically selecting a model. If the receptionist knows which model to use for the applicant, she selects one of the displayed models, otherwise she clicks a button. For example, when the receptionist determines that the applicant is displeased, he selects a model associated with weight information that includes a condition specifying displeased for the characteristic "voice tone."
　ステップＳ２３では、選択部１２２は、モデル記憶部１４に格納されているモデルの中から、ステップＳ２１において取得された申込者情報に対応するモデルを選択する。具体的には、選択部１２２は、モデル記憶部１４に対して申込者情報に含まれる特性情報に合致する重み情報の探索を行い、モデル記憶部１４から特性情報に合致する重み情報に関連付けられているモデルを読み出す。 In step S23, the selection unit 122 selects the model corresponding to the applicant information acquired in step S21 from among the models stored in the model storage unit 14. Specifically, the selection unit 122 searches the model storage unit 14 for weight information that matches the characteristic information included in the applicant information, and associates weight information that matches the characteristic information from the model storage unit 14 with the weight information that matches the characteristic information. Read the model that is
　一方、ステップＳ２４では、選択部１２２は、受付担当者の指示に従って、モデル記憶部１４に格納されているモデルのうちの１つのモデルを選択する。例えば、選択部１２２は、モデル記憶部１４から受付担当者により選択されたモデルを読み出す。 On the other hand, in step S24, the selection unit 122 selects one of the models stored in the model storage unit 14 according to the instructions of the person in charge of reception. For example, the selection unit 122 reads the model selected by the receptionist from the model storage unit 14 .
　ステップＳ２５において、予測部１２３は、ステップＳ２３又はＳ２４において選択されたモデルを使用して、申込者情報に基づいて申込者の希望工事日を予測する。予測部１２３は、申込者情報に含まれる申込者の特性情報と候補日情報とをモデルに入力し、モデルから候補日のそれぞれに対するスコアを得る。予測部１２３は、スコアが最も高い候補日を第１候補として特定してよい。予測部１２３は、スコアが２番目に高い候補日を第２候補としてさらに特定してよく、スコアが３番目に高い候補日を第３候補としてさらに特定してよい。 In step S25, the prediction unit 123 uses the model selected in step S23 or S24 to predict the applicant's desired construction date based on the applicant information. The prediction unit 123 inputs the applicant's characteristic information and candidate date information included in the applicant information into the model, and obtains a score for each candidate date from the model. The prediction unit 123 may identify the candidate date with the highest score as the first candidate. The prediction unit 123 may further identify the candidate date with the second highest score as the second candidate, and may further identify the candidate date with the third highest score as the third candidate.
　ステップＳ２６において、出力部１２４は、希望工事日の予測結果を出力する。例えば、出力部１２４は、希望工事日の予測結果を表示装置に表示する。例えば、希望工事日の予測結果は、第１候補として特定された候補日を示す情報と、他の候補日のそれぞれに対するスコアと、を含んでよい。代替として、希望工事日の予測結果は、全ての候補日のそれぞれに対するスコアを含んでよい。また、希望工事日の予測結果は、第１候補として特定された候補日、第２候補として特定された候補日、第３候補として特定された候補日を示す情報を含んでよい。 In step S26, the output unit 124 outputs the prediction result of the desired construction date. For example, the output unit 124 displays the prediction result of the desired construction date on the display device. For example, the prediction result of the desired construction date may include information indicating the candidate date identified as the first candidate and scores for each of the other candidate dates. Alternatively, the desired construction date prediction results may include a score for each of all candidate dates. Moreover, the prediction result of the desired construction date may include information indicating the candidate date specified as the first candidate, the candidate date specified as the second candidate, and the candidate date specified as the third candidate.
　［効果］
　情報処理装置１０は、予約者の少なくとも１つの特性に対する重みを示す重み情報を取得し、複数の予約者に関する予約状況データの中から、重み情報に対応する予約状況データを抽出し、抽出された予約状況データを学習データとして使用して、申込者の希望工事日を予測するためのモデルを学習する。モデルは、申込者の特性情報と光回線工事の候補日を示す候補日情報とを入力とし、候補日のそれぞれに対するスコアを出力するように構成される。スコアは、光回線工事が候補日に実施されることに対する申込者の想定満足度を表す。情報処理装置１０は、抽出された予約状況データに対して光回線工事の予約を自由に実行できると仮定した場合に得られる工事日の分布において頻度が多い日ほどスコアが高くなるように、モデルを学習する。モデルは、物又はサービスに対する満足の度合いを数値に置き換える関数である効用関数に相当する。このようにして生成されたモデルは、光回線工事の予約を申し込む申込者に対して工事日の提案を効率的に行うことを可能にする。 [effect]
The information processing device 10 acquires weight information indicating a weight for at least one characteristic of a reservation person, extracts reservation status data corresponding to the weight information from reservation status data related to a plurality of reservation persons, and extracts the extracted reservation status data. Using the reservation status data as learning data, train a model to predict the applicant's desired construction date. The model is configured to receive as input characteristic information of applicants and candidate date information indicating candidate dates for optical line installation, and to output scores for each of the candidate dates. The score represents the applicant's assumed degree of satisfaction with the construction of the optical fiber line on the candidate date. The information processing apparatus 10 sets the model so that the more frequent the day, the higher the score in the distribution of work days obtained when it is assumed that the reservation for the optical line work can be freely executed for the extracted reservation status data. to learn. A model corresponds to a utility function, which is a function that converts the degree of satisfaction with a product or service into a numerical value. The model generated in this way makes it possible to efficiently propose a construction date to applicants who apply for reservations for optical line construction.
　少なくとも１つの特性は声色を含んでよい。これにより、電話を用いた予約システムにおいて、工事日の提案をより適切に行うことが可能になる。 At least one characteristic may include timbre. This makes it possible to more appropriately propose a construction date in a telephone-based reservation system.
　情報処理装置１０は、複数のモデルを重み情報に関連付けてモデル記憶部１４に格納し、申込者の特性を示す特性情報を取得し、複数のモデルの中からモデルを選択し、申込者の特性情報と候補日情報とを選択されたモデルに入力し、選択されたモデルから出力されるスコアに基づいて、申込者の希望工事日を予測し、予測結果を出力する。情報処理装置１０は、複数のモデルの中から、申込者の特性情報に対応する重み情報に関連付けられているモデルを選択してよい。受付担当者は、申込者に対して工事日を提案するために、情報処理装置１０から出力される予測結果を参照する。情報処理装置１０によれば、受付担当者による工事日の提案を効率化することが可能となる。さらに、申込者が合意した工事日におけるスコアにより、申込者の満足度を推定することができる。例えば、申込者が合意した工事日におけるスコアが閾値より低い場合に、申込者の満足度が低いと推定することができる。その場合、受付担当者は、付加サービスを提示したり、より丁寧に対応したりするなど、何らかのアクションを取ることができる。候補日のそれぞれに対するスコア（想定満足度）は、申込者の希望が実現できない場合において、付加サービスを提示すべきか否かや、より丁寧な対応が必要となるか否かの判断材料として使用することができる。 The information processing apparatus 10 associates a plurality of models with weight information and stores them in the model storage unit 14, acquires characteristic information indicating characteristics of the applicant, selects a model from among the plurality of models, and stores the characteristics of the applicant. Information and candidate date information are input to the selected model, based on the score output from the selected model, the applicant's desired construction date is predicted, and the prediction result is output. The information processing device 10 may select a model associated with weight information corresponding to the applicant's characteristic information from among a plurality of models. The person in charge of reception refers to the prediction result output from the information processing device 10 in order to propose the construction date to the applicant. According to the information processing device 10, it is possible to make the proposal of the construction date by the person in charge of reception more efficient. Furthermore, the degree of satisfaction of the applicant can be estimated from the score on the construction date agreed upon by the applicant. For example, it can be inferred that the applicant's satisfaction level is low if the score on the construction date agreed upon by the applicant is lower than a threshold. In that case, the receptionist can take some action, such as suggesting additional services or responding more politely. The score (expected satisfaction) for each candidate date is used as a criterion for deciding whether additional services should be offered or whether a more polite response is required when the applicant's wishes cannot be realized. be able to.
　ホテルや航空券などの予約業務では、効用関数を推察するために有効な指標として価格を利用することができる。顧客ごとに各日で差がある価格を提示し、顧客による選択というフィードバックを得ることにより、顧客ごとの効用関数を推察することが可能である。 In reservation operations such as hotels and airline tickets, price can be used as an effective indicator for inferring the utility function. It is possible to infer the utility function for each customer by presenting different prices to each customer on each day and obtaining feedback of customer selection.
　一方、通信業界では、約款により各種サービスに対する価格が決まっており、効用関数を推察するために有効な指標を利用することができない。さらに、各サービス実施候補日（例えば空き工事日）に対する満足度が不明確であり、予約が完了したとしてもそれが妥協の末なのか充分満足した結果なのかという、今後のサービス提供を検討するうえで必要になる情報の推察が難しい。 On the other hand, in the telecommunications industry, the prices for various services are determined by the terms and conditions, and it is not possible to use effective indicators to infer the utility function. In addition, the degree of satisfaction with each service implementation candidate date (for example, a vacant construction date) is unclear, and even if the reservation is completed, consider whether it is the end of compromise or the result of being fully satisfied, and consider future service provision. It is difficult to infer the information that will be needed.
　本実施形態では、価格などの効用関数を推察するために有効な指標を利用できることを必要としない。このため、本実施形態は、上記のような通信業界におけるサービスに対する予約にも適用することができる。明確な希望工事日を持たない申込者に対しても、満足感の高い工事日を提示できる可能性が高まり、提示までの時間を短縮することが可能となる。 In this embodiment, it is not necessary to be able to use effective indicators to infer utility functions such as prices. Therefore, this embodiment can also be applied to reservations for services in the communication industry as described above. Even for an applicant who does not have a definite desired construction date, the possibility of being able to present a highly satisfying construction date increases, and it is possible to shorten the time until the proposal is made.
　なお、本発明は、上記実施形態に限定されるものではなく、実施段階ではその要旨を逸脱しない範囲で種々に変形することが可能である。また、各実施形態は適宜組み合わせて実施してもよく、その場合組み合わせた効果が得られる。さらに、上記実施形態には種々の発明が含まれており、開示される複数の構成要素から選択された組み合わせにより種々の発明が抽出され得る。例えば、実施形態に示される全構成要素からいくつかの構成要素が削除されても、課題が解決でき、効果が得られる場合には、この構成要素が削除された構成が発明として抽出され得る。 It should be noted that the present invention is not limited to the above-described embodiments, and can be variously modified in the implementation stage without departing from the gist of the present invention. Further, each embodiment may be implemented in combination as appropriate, in which case the combined effect can be obtained. Furthermore, various inventions are included in the above embodiments, and various inventions can be extracted by combinations selected from the disclosed plurality of components. For example, even if some components are deleted from all the components shown in the embodiment, if the problem can be solved and effects can be obtained, the configuration in which these components are deleted can be extracted as an invention.
　１０…情報処理装置
　１１…モデル生成部
　１１１…取得部
　１１２…抽出部
　１１３…学習部
　１２…情報提示部
　１２１…取得部
　１２２…選択部
　１２３…予測部
　１２４…出力部
　１３…予約状況記憶部
　１４…モデル記憶部
　５１…プロセッサ
　５２…ＲＡＭ
　５３…プログラムメモリ
　５４…ストレージデバイス
　５５…入出力インタフェース
  DESCRIPTION OF SYMBOLS 10... Information processing apparatus 11... Model generation part 111... Acquisition part 112... Extraction part 113... Learning part 12... Information presentation part 121... Acquisition part 122... Selection part 123... Prediction part 124... Output part 13... Reservation status storage part 14 ... model storage unit 51 ... processor 52 ... RAM
53 Program memory 54 Storage device 55 Input/output interface

Claims (7)

  1. 　予約者の少なくとも１つの特性に対する重みを示す重み情報を取得する第１の取得部と、
    　複数の予約者に関する予約状況データの中から、前記重み情報に対応する予約状況データを抽出する抽出部であって、前記複数の予約者の各々に関する予約状況データは、前記予約者について完了済みの予約の内容を示し、前記予約者の特性を示す特性情報を含む、抽出部と、
    　前記抽出された予約状況データを学習データとして使用して、申込者の特性を示す特性情報とサービス実施の候補日を示す候補日情報とを入力とし、前記候補日のそれぞれに対する、サービスが候補日に実施されることに対する前記申込者の想定満足度を表すスコアを出力するモデルを学習する学習部と、
    　を備える情報処理装置。     a first acquisition unit that acquires weight information indicating a weight for at least one characteristic of a reservation person;
    an extracting unit for extracting reservation status data corresponding to the weight information from reservation status data about a plurality of reserving persons, wherein the reservation status data about each of the plurality of reserving persons is a complete reservation status data for each of the reserving persons; an extraction unit that indicates the content of the reservation and includes property information that indicates the property of the person making the reservation;
    Characteristic information indicating characteristics of the applicant and candidate date information indicating candidate dates for service implementation are input using the extracted reservation status data as learning data, and the service is provided on the candidate date for each of the candidate dates. a learning unit that learns a model that outputs a score representing the applicant's expected satisfaction with being performed in
    Information processing device.
  2. 　前記学習部は、前記抽出された予約状況データに対してサービスの予約を自由に実行できると仮定した場合に得られるサービス実施日の分布において頻度が多い日ほどスコアが高くなるように、前記モデルを学習する、
    　請求項１に記載の情報処理装置。     The learning unit calculates the model so that the more frequent the day in the service implementation date distribution obtained when it is assumed that service reservations can be freely made for the extracted reservation status data, the higher the score. to learn the
    The information processing device according to claim 1 .
  3. 　前記少なくとも１つの特性は声色を含む、請求項１又は２に記載の情報処理装置。 The information processing apparatus according to claim 1 or 2, wherein said at least one characteristic includes voice tone.
  4. 　前記学習部により得られる複数のモデルを重み情報に関連付けて格納するモデル記憶部と、
    　申込者の特性を示す特性情報を取得する第２の取得部と、
    　前記複数のモデルの中からモデルを選択する選択部と、
    　前記申込者の前記特性情報と候補日情報とを前記選択されたモデルに入力し、前記選択されたモデルから出力されるスコアに基づいて、前記申込者がサービスの実施を希望する希望サービス実施日を予測する予測部と、
    　前記希望サービス実施日の予測結果を出力する出力部と、
    　をさらに備える請求項１乃至３のいずれか１項に記載の情報処理装置。     a model storage unit that stores a plurality of models obtained by the learning unit in association with weight information;
    a second acquisition unit that acquires characteristic information indicating characteristics of an applicant;
    a selection unit that selects a model from among the plurality of models;
    The applicant's characteristic information and candidate date information are input to the selected model, and the desired service implementation date on which the applicant wishes to perform the service based on the score output from the selected model. a prediction unit that predicts
    an output unit that outputs a prediction result of the desired service implementation date;
    The information processing apparatus according to any one of claims 1 to 3, further comprising:
  5. 　前記選択部は、前記複数のモデルの中から、前記申込者の前記特性情報に対応する重み情報に関連付けられているモデルを選択する、請求項４に記載の情報処理装置。 The information processing apparatus according to claim 4, wherein the selection unit selects, from among the plurality of models, a model associated with weight information corresponding to the characteristic information of the applicant.
  6. 　予約者の少なくとも１つの特性に対する重みを示す重み情報を取得することと、
    　複数の予約者に関する予約状況データの中から、前記重み情報に対応する予約状況データを抽出することであって、前記複数の予約者の各々に関する予約状況データは、前記予約者について完了済みの予約の内容を示し、前記予約者の特性を示す特性情報を含む、ことと、
    　前記抽出された予約状況データを学習データとして使用して、申込者の特性を示す特性情報とサービス実施の候補日を示す候補日情報とを入力とし、前記候補日のそれぞれに対する、サービスが候補日に実施されることに対する前記申込者の想定満足度を表すスコアを出力するモデルを学習することと、
    　を備える情報処理方法。     obtaining weight information indicating a weight for at least one characteristic of the subscriber;
    extracting reservation status data corresponding to the weight information from among reservation status data relating to a plurality of reserving persons, wherein the reservation status data relating to each of the plurality of reserving persons includes completed reservations for the reserving persons; and including characteristic information indicating the characteristics of the reservation person;
    Characteristic information indicating characteristics of the applicant and candidate date information indicating candidate dates for service implementation are input using the extracted reservation status data as learning data, and the service is provided on the candidate date for each of the candidate dates. learning a model that outputs a score representing the applicant's perceived satisfaction with being performed in
    An information processing method comprising:
  7. 　請求項１乃至５のいずれか１項に記載の情報処理装置が備える各部としてコンピュータを機能させるためのプログラム。
          A program for causing a computer to function as each unit included in the information processing apparatus according to any one of claims 1 to 5.
PCT/JP2021/018489 2021-05-14 2021-05-14 Information processing device, information processing method, and program WO2022239246A1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/JP2021/018489 WO2022239246A1 (en) 2021-05-14 2021-05-14 Information processing device, information processing method, and program

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/JP2021/018489 WO2022239246A1 (en) 2021-05-14 2021-05-14 Information processing device, information processing method, and program

Publications (1)

Publication Number Publication Date
WO2022239246A1 true WO2022239246A1 (en) 2022-11-17

Family

ID=84028081

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2021/018489 WO2022239246A1 (en) 2021-05-14 2021-05-14 Information processing device, information processing method, and program

Country Status (1)

Country Link
WO (1) WO2022239246A1 (en)

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2005062240A (en) * 2003-08-13 2005-03-10 Fujitsu Ltd Audio response system
JP2007060622A (en) * 2005-03-29 2007-03-08 Matsushita Electric Ind Co Ltd Network electric household appliance system and its program
JP2017228221A (en) * 2016-06-24 2017-12-28 トヨタ自動車株式会社 Reservation device, reservation method and on-vehicle system

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2005062240A (en) * 2003-08-13 2005-03-10 Fujitsu Ltd Audio response system
JP2007060622A (en) * 2005-03-29 2007-03-08 Matsushita Electric Ind Co Ltd Network electric household appliance system and its program
JP2017228221A (en) * 2016-06-24 2017-12-28 トヨタ自動車株式会社 Reservation device, reservation method and on-vehicle system

Similar Documents

Publication Publication Date Title
US11019210B2 (en) Customer journey management
US9509846B1 (en) Systems and methods of natural language processing to rank users of real time communications connections
JP7043786B2 (en) Sales activity support system, sales activity support method and sales activity support program
JP7480386B2 (en) Automated bot creation based on scripts
JP6848883B2 (en) Information processing equipment, information processing methods, and programs
US20220172254A1 (en) Automated deal guide structure identification
US20210065223A1 (en) Survey data collection, analysis, and categorization systems and methods
US9336212B2 (en) Systems and methods for collection and automatic analysis of opinions on various types of media
CN111028007A (en) User portrait information prompting method, device and system
US20140195312A1 (en) System and method for management of processing workers
US20190052756A1 (en) Systems and methods to enroll users for real time communications connections
JP2020126392A (en) Selection device, selection method, and selection program
WO2022239246A1 (en) Information processing device, information processing method, and program
US20200226641A1 (en) Automated deal guide optimization
US20190318384A1 (en) Automated deal guide user interface
CN113536103A (en) Information recommendation method and device, electronic equipment and storage medium
JP7527063B1 (en) Management Device
CN110414824B (en) Service management system based on community system
WO2012131993A1 (en) Information providing apparatus, information providing method, information providing program, and computer-readable recording medium having information providing program recorded therein
JP2022069602A (en) Sales activity support system, sales activity support method and sales activity support program
CN113869932A (en) Client screening method, device, equipment and storage medium
JP2005092818A (en) Commodity development support system, method, and program

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: 21941976

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 21941976

Country of ref document: EP

Kind code of ref document: A1