System for guiding an operator to question a rescue sender

Abstract

Embodiments of the present application provide a system for guiding an operator to question a rescue sender. The system uses the positive example probability of any second reference group included in any first reference group among a plurality of first reference groups as a cut-off point to perform a classification test on any first reference group to obtain the classification sensitivity and classification specificity corresponding to any second reference group, and based on the classification sensitivity and classification specificity corresponding to a plurality of second reference groups within any first reference group, obtain a receiver operating characteristic curve corresponding to any first reference group. Based on the receiver operating characteristic curves corresponding to each first reference group within a plurality of first reference groups, determine a target first reference group from the plurality of first reference groups, and determine target question semantics from the distance between the coordinate point corresponding to any second reference group among the plurality of second reference groups and the perfect classification coordinate point.

Description

Technical Field

[0001] (Cross - reference to related applications) This application claims priority to Chinese Patent Application No. 202210805597.1, titled "SYSTEM FOR HELPING OPERATOR TO QUESTION HELP - SEEKER", filed with the China National Intellectual Property Administration on July 8, 2022, and incorporates its entire content by reference.

[0002] This application relates to the technical field of computers, and in particular, to a system for guiding an operator to question a help - seeker. This application also relates to a system for guiding an operator to determine whether a person in need of help is a positive example. This application further relates to a system for guiding an investigator to question an interviewee. This application further relates to a system for guiding an investigator to determine whether an interviewee is a positive example.

Background Art

[0003] Out - of - hospital cardiac arrest (OHCA) is the most severe clinical condition that occurs outside the hospital, and once it occurs, the patient may die immediately. The golden time for rescue after cardiac arrest is only 3 - 5 minutes, but the average response time of ambulances in China is usually more than 15 minutes, and on - site cardiopulmonary resuscitation (CPR) is the only hope for the survival of OHCA patients.

[0004] In recent years, attempts have been made at home and abroad for operators at emergency medical dispatch centers (e.g., emergency phone number: 120) to encourage and guide the rescuer (the caller) or other people present at the scene to provide on-site CPR (i.e., phone CPR) to OHCA patients when receiving a distress call. For operators lacking in-depth emergency response expertise, the implementation rate and success rate of phone CPR are low, and it is easy to overlook the diagnosis of OHCA patients. Therefore, how to guide the operator to question the rescuer, and how to provide reference data to the operator to determine the on-site situation of the victim (e.g., whether the victim is suspected of being an OHCA patient) and the progress of the rescue of the victim by the rescuer or other personnel present at the scene (e.g., whether the rescuer has started CPR, whether an automated external defibrillator (AED) has been used, whether return of spontaneous circulation (ROSC) has been achieved after CPR, etc.) are technical problems that need to be solved urgently.

Summary of the Invention

Problems to be Solved by the Invention

[0005] Embodiments of the present application provide a system for guiding an operator to question the rescuer, guiding the operator to question the rescuer, and assisting the operator to determine the on-site situation of the victim (e.g., whether the victim is suspected of being an OHCA patient), and the progress of the rescue of the victim by the rescuer or other personnel present at the scene (e.g., whether the rescuer has started CPR, whether an AED has been used, whether return of spontaneous circulation (ROSC) has been achieved after CPR, etc.). Embodiments of the present application also provide a system for determining the positive example probability of the victim, assisting the operator to determine the on-site situation of the victim (e.g., whether the victim is suspected of being an OHCA patient, etc.), and the progress of the rescue of the victim by the rescuer or other personnel present at the scene (e.g., whether the rescuer has started CPR, whether an AED has been used, whether return of spontaneous circulation (ROSC) has been achieved after CPR, etc.).

Means for Solving the Problems

[0006] One embodiment of the present application provides a system for guiding an operator to question a rescue caller. The system includes a first reference group acquisition module, a classification test module, a receiver operating characteristic curve acquisition module, a candidate question semantics acquisition module, and a target question semantics acquisition module. The first reference group acquisition module is configured to select a plurality of unprocessed first reference groups from an initial reference group. Here, the first reference group is obtained by classifying a plurality of historical rescued persons in the initial reference group based on entity attribute clusters. The entity attribute cluster is composed of a plurality of entity attribute pairs corresponding to the same core question. The entity attribute pair includes an entity and an attribute. The entity represents the semantics of the operator's question in a historical rescue call, and the attribute represents the semantics of the rescue caller's answer to the question. The classification test module is configured to use the positive example probability of any second reference group included in any first reference group among the plurality of first reference groups as a cut-off point to perform a classification test on the any first reference group to obtain the classification sensitivity and classification specificity corresponding to the any second reference group. Here, the cut-off point is for determining the positive example label of the first reference group, and the second reference group is obtained by classifying the first reference group based on entity attributes. The receiver operating characteristic curve acquisition module is configured to obtain the receiver operating characteristic curve corresponding to the any first reference group based on the classification sensitivity and classification specificity corresponding to the plurality of second reference groups within the any first reference group. The candidate question semantics acquisition module is configured to determine a target first reference group from the plurality of first reference groups based on the receiver operating characteristic curve corresponding to each first reference group within the plurality of first reference groups, and use the entity of each entity attribute pair within the entity attribute cluster corresponding to the target first reference group as candidate question semantics. The target question semantics acquisition module isBased on the distance from the coordinate point corresponding to any second reference group among the plurality of second reference groups in the receiver operating characteristic curve corresponding to the target first reference group to the perfect classification coordinate point, it is configured to determine the target question semantics from the candidate question semantics. Here, the perfect classification coordinate point is the coordinate point where the abscissa in the receiver operating characteristic curve is 0 and the ordinate is 1, and the target question semantics is configured to guide the operator to ask the next question.,

[0007] In a plurality of embodiments, taking the positive example probability of any second reference group included in any first reference group among the plurality of first reference groups as a cut-off point and performing a classification test on the any first reference group to obtain the classification sensitivity and classification specificity corresponding to the any second reference group is Regarding the members of the second reference group whose positive example probability in the first reference group is greater than or equal to the cut-off point as preset positive examples, and regarding the members of the second reference group whose positive example probability in the first reference group is less than the cut-off point as preset negative examples, obtaining the classification sensitivity according to the number of members of the preset positive examples actually classified as positive examples and the number of members of the first reference group actually classified as positive examples, and obtaining the classification specificity according to the number of members of the preset negative examples actually classified as negative examples and the number of members of the first reference group actually classified as negative examples.

[0008] In a plurality of embodiments, obtaining a receiver operating characteristic curve corresponding to any first reference group based on classification sensitivities and classification specificities corresponding to a plurality of second reference groups within the any first reference group includes: taking the classification sensitivity of any second reference group among the plurality of second reference groups as the vertical coordinate, taking the absolute value of the difference between the classification specificity of the any second reference group and 1 as the horizontal coordinate, and obtaining a coordinate point corresponding to the any second reference group; and connecting the coordinate points corresponding to each second reference group within the first reference group to obtain a receiver operating characteristic curve corresponding to the any first reference group.

[0009] In a plurality of embodiments, determining a target first reference group from the plurality of first reference groups based on the receiver operating characteristic curves corresponding to each first reference group within the plurality of first reference groups includes: calculating the area under the curve of the receiver operating characteristic curve corresponding to each first reference group within the plurality of first reference groups; and using the first reference group corresponding to the maximum area under the curve as the target first reference group.

[0010] In a plurality of embodiments, determining target query semantics from the candidate query semantics based on the distance from the coordinate point corresponding to any second reference group among the plurality of second reference groups in the receiver operating characteristic curve corresponding to the target first reference group to the perfect classification coordinate point includes: calculating a plurality of distances from the coordinate point corresponding to each second reference group among the plurality of second reference groups to the perfect classification coordinate point; and using the second reference group corresponding to the shortest distance among the plurality of distances as the target second reference group, and using the entity of the entity-attribute pair corresponding to the target second reference group as the target query semantics.

[0011] In a plurality of embodiments, personal characteristic information of the rescued person is extracted from the information spoken by the rescuer, and the initial reference group is obtained by determining an initial reference group based on the personal characteristic information.

[0012] One embodiment of the present application provides a system for guiding an operator to determine whether a person in need of assistance is a positive example, and the system includes an entity attribute clustering and entity attribute pair determination module, a first reference group and a second reference group determination module, a preset positive example member number acquisition module, and a positive example probability determination module.The entity attribute cluster and entity attribute pair determination module are configured to determine an entity attribute cluster currently corresponding to the rescued person and an entity attribute pair currently corresponding to the rescued person based on a question of the operator and an answer of the rescued person to the question in any interaction between the operator and the rescue caller, where the entity attribute cluster is composed of a plurality of entity attribute pairs corresponding to the same core question, the entity attribute pair includes an entity and an attribute, the entity represents the semantics of the operator's question in the historical rescue call, the attribute represents the semantics of the rescued person's answer to the question, the first reference group and second reference group determination module is configured to obtain a first reference group corresponding to the entity attribute cluster and a second reference group corresponding to the entity attribute pair from an initial reference group, where the first reference group is obtained by classifying a plurality of historical rescued persons in the initial reference group based on the entity attribute cluster, the second reference group is obtained by classifying the first reference group based on the entity attribute, the preset positive example member number acquisition module is configured to obtain the number of members of the first reference group that are preset positive examples using the positive example probability of the second reference group as a cut-off point, where the cut-off point is for determining the positive example label of the first reference group, the positive example probability determination module is configured to obtain the positive example probability of the rescued person based on the ratio of the number of members actually classified as positive examples among the members of the first reference group that are preset positive examples to the number of members that are preset positive examples in the first reference group, and the positive example probability is configured to guide the operator to determine whether the rescued person is a positive example.

[0013] In a plurality of embodiments, the system further includes a positive example probability trend scatter diagram acquisition module and a slope acquisition module for a linear regression equation. The positive example probability trend scatter diagram acquisition module is configured to acquire a positive example probability trend scatter diagram based on a plurality of positive example probabilities of the person in need of rescue obtained in a plurality of conversations between the operator and the rescue caller. The slope acquisition module for the linear regression equation is configured to fit the positive example probability trend scatter diagram using a linear regression equation and solve the linear regression equation to obtain the slope of the linear regression equation. Here, the positive example probability trend scatter diagram and the slope of the linear regression equation are for representing the trend of the probability that the person in need of rescue is a positive example.

[0014] In a plurality of embodiments, acquiring a positive example probability trend scatter diagram based on a plurality of positive example probabilities of the person in need of rescue obtained in a plurality of conversations between the operator and the rescue caller includes obtaining coordinates on the positive example probability trend scatter diagram with the number of conversations corresponding to any one of the plurality of positive example probabilities as the abscissa and the any one positive example probability as the ordinate. In a plurality of embodiments, determining the entity attribute cluster currently corresponding to the person in need of rescue and the entity attribute pair currently corresponding to the person in need of rescue includes obtaining the operator's question, extracting question semantics information from the question, obtaining the rescue caller's answer to the question, extracting answer semantics information from the answer, determining the core question described in the question, obtaining the entity attribute cluster based on the core question, and determining the entity attribute pair based on the question semantics information and the answer semantics information.

[0015] In a plurality of embodiments, taking the positive example probability of the second reference group as a cut-off point and obtaining the number of members of the first reference group that are preset positive examples includes taking the number of members of the second reference group whose positive example probability in the first reference group is greater than or equal to the cut-off point as the number of members of the preset positive examples.

[0016] In a plurality of embodiments, the system further includes a rate ratio acquisition module and a rate difference acquisition module, the rate ratio acquisition module is configured to acquire a rate ratio based on the ratio of the current positive example probability of the rescued person to the positive example probability of the rescued person obtained in the previous conversation, and the rate difference acquisition module is configured to acquire a rate difference according to the difference between the current positive example probability of the rescued person and the positive example probability of the rescued person obtained in the previous conversation. Here, the rate difference and the rate ratio are configured to guide the operator to determine the value of whether the current rescued person is a positive example based on the current positive example probability of the rescued person.

[0017] One embodiment of the present application provides a system for guiding an investigator to ask questions to an interviewee. The system includes a first reference group acquisition module, a classification test module, a receiver operating characteristic curve acquisition module, a candidate question semantics acquisition module, and a target question semantics acquisition module. The first reference group acquisition module is configured to select a plurality of unprocessed first reference groups from an initial reference group. Here, the first reference group is obtained by classifying a plurality of historical interviewees in the initial reference group based on entity attribute clusters. The entity attribute cluster is composed of a plurality of entity attribute pairs corresponding to the same core question. The entity attribute pair includes an entity and an attribute. The entity represents the semantics of the investigator's question in a historical survey call, and the attribute represents the semantics of the interviewee's answer to the question. The classification test module is configured to use the positive example probability of any second reference group included in any first reference group among the plurality of first reference groups as a cut-off point to perform a classification test on any first reference group to obtain the classification sensitivity and classification specificity corresponding to any second reference group. Here, the cut-off point is for determining the positive example label of the first reference group, and the second reference group is obtained by classifying the first reference group based on entity attributes. The receiver operating characteristic curve acquisition module is configured to obtain a receiver operating characteristic curve corresponding to any first reference group based on the classification sensitivity and classification specificity corresponding to a plurality of second reference groups within any first reference group. The candidate question semantics acquisition module is configured to determine a target first reference group from the plurality of first reference groups based on the receiver operating characteristic curve corresponding to each first reference group within the plurality of first reference groups, and use the entity of each entity attribute pair within the entity attribute cluster corresponding to the target first reference group as candidate question semantics. The target question semantics acquisition module isBased on the distance from the coordinate point corresponding to any second reference group among the plurality of second reference groups in the receiver operating characteristic curve corresponding to the target first reference group to the perfect classification coordinate point, it is configured to determine the target question semantics from the candidate question semantics. Here, the perfect classification coordinate point is the coordinate point where the abscissa in the receiver operating characteristic curve is 0 and the ordinate is 1, and the target question semantics is configured to guide the investigator to ask the following question.,

[0018] One embodiment of the present application provides a system for guiding an investigator to determine whether an interviewee is a positive example. The system includes an entity attribute cluster and entity attribute pair determination module, a first reference group and second reference group determination module, a preset positive example member number acquisition module, and a positive example probability determination module. The entity attribute cluster and entity attribute pair determination module is configured to determine an entity attribute cluster currently corresponding to the interviewee and an entity attribute pair currently corresponding to the interviewee based on a question of the investigator and the interviewee's answer to the question in any conversation between the investigator and the interviewee. Here, the entity attribute cluster is composed of a plurality of entity attribute pairs corresponding to the same core question. The entity attribute pair includes an entity and an attribute. The entity represents the semantics of the investigator's question in a historical survey call, and the attribute represents the semantics of the interviewee's answer to the question. The first reference group and second reference group determination module is configured to obtain a first reference group corresponding to the entity attribute cluster and a second reference group corresponding to the entity attribute pair from an initial reference group. Here, the first reference group is obtained by classifying a plurality of historical interviewees into the initial reference group based on the entity attribute cluster, and the second reference group is obtained by classifying the first reference group based on the entity attribute. The preset positive example member number acquisition module is configured to obtain the number of members of the first reference group that are preset positive examples with the positive example probability of the second reference group as a cut-off point. Here, the cut-off point is for determining the positive example label of the first reference group. The positive example probability determination module is configured to obtain the positive example probability of the interviewee based on the ratio of the number of members actually classified as positive examples among the members of the first reference group that are preset positive examples to the number of members that are preset positive examples in the first reference group. The positive example probability is configured to guide the investigator to determine whether the interviewee is a positive example. [Effect of the Invention]

[0019] In the embodiment provided by the present application, the entity attribute cluster is composed of entity attribute pairs having the same theme in the historical rescue call. The entity attribute pair is composed of the operator's question semantics and the rescuer's answer semantics, structuring a large number of chaotic historical conversations in the rescue dispatch database, thereby classifying the historical rescuer targets by the historical conversation to obtain a first reference group and a second reference group. The embodiment provided by the present application uses the positive example probability of any second reference group as a cut-off point to perform a classification test on the first reference group, making the second reference group obtained by classification with entity attributes continuously separable, making the first reference group obtained by classification based on the entity attribute cluster computable. Furthermore, the classification sensitivity and classification specificity obtained with the positive example probability of the second reference group as a cut-off point can more accurately reflect the effectiveness of the entity attribute pair corresponding to the second reference group in determining the on-site situation of the rescued person and the progress of the rescuer's rescue of the rescued person through the represented conversation.

[0020] In the embodiment provided by the present application, a receiver operating characteristic curve corresponding to the first reference group is obtained based on the classification sensitivity and classification specificity corresponding to any second reference group, quantifying the effectiveness of multiple entity attribute clusters when determining the on-site situation of the rescued person and the progress of the rescuer's rescue of the rescued person. In this way, the core questions of the entity attribute cluster representation useful for determining the on-site situation of the rescued person and the progress of the rescuer's rescue of the rescued person can be obtained.

[0021] In the embodiments provided by this application, by calculating the distance from the coordinate point corresponding to any second reference group among a plurality of second reference groups in the receiver operation characteristic curve corresponding to the obtained target first reference group to the perfect classification coordinate point, the effectiveness of the entity attribute pair in determining the situation at the scene of the rescued person and the progress of the rescue from the rescuer to the rescued person is quantified. In this way, the question semantics of the entity attribute pair representation that helps to determine the situation at the scene of the rescued person and the progress of the rescue from the rescuer to the rescued person can be obtained.

[0022] In the embodiments provided by this application, through the operator's question and the rescuer's answer to the question, the entity attribute cluster currently corresponding to the rescued person and the entity attribute pair currently corresponding to the rescued person are determined, and the first reference group corresponding to the entity attribute cluster and the second reference group corresponding to the entity attribute pair are obtained from the initial reference group. In this way, the historical reference group of the rescued person can be obtained through the conversation. In the embodiments provided by this application, taking the positive example probability of the second reference group as the cut-off point, the number of members of the first reference group that is the preset positive example is obtained. Based on the ratio between the number of members actually classified as positive examples among the members of the first reference group that is the preset positive example and the number of members that are preset positive examples in the first reference group, the positive example probability of the rescued person is obtained. Thereby, the patient record resources in the rescue dispatch database are utilized to the maximum extent, and through the conversation between the operator and the rescuer, the probability of whether the rescued person is a positive example can be obtained more accurately.

Brief Description of the Drawings

[0023] This application is further illustrated by exemplary embodiments, and these exemplary embodiments are described in detail by the drawings. These embodiments are not limiting. In these embodiments, the same numbers represent the same structures.

[0024]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Modes for Carrying Out the Invention

[0025] To more clearly explain the technical solutions of the embodiments of the present application, the drawings necessary for the description of the embodiments are briefly introduced below. Obviously, the drawings in the following description are only multiple examples or embodiments of the present application, and those skilled in the art can also apply the present application to other similar scenarios based on these drawings without creative efforts. Unless it is obvious from the language environment or there is a separate description, the same reference numerals in the drawings represent the same structure or operation.

[0026] As used herein, the terms "system," "apparatus," "unit," and / or "module" are understood to be a means for distinguishing different components, elements, parts, portions, or assemblies at different levels. However, they can be replaced with other expressions when the same purpose can be achieved.

[0027] As shown in this application and the claims, unless the context clearly indicates otherwise, words such as "one," "a," "an," and / or "the" do not particularly refer to the singular form and may include the plural form. Generally speaking, the terms "comprising" and "including" only include the clearly identified steps and elements, and these steps and elements do not constitute an exclusive listing, and other steps or elements may be included in the method or device.

[0028] In this application, flowcharts are used to describe the operations performed by the system according to the embodiments of this application. It should be understood that the operations before and after are not necessarily executed in the exact order. Instead, these steps can be processed in the reverse order or simultaneously. At the same time, other operations can be added to these processes, or one or more steps can be deleted from these processes.

[0029] FIG. 1 is a schematic diagram of an application scenario of a system that guides an operator to question a rescue sender and a system that guides an operator to determine whether a rescued person is a positive example according to some embodiments of this application.

[0030] The systems provided by the embodiments of the present application, which guide the operator to question the rescue sender and guide the operator to determine whether the rescued person is a positive example, can be applied to various task scenarios, such as the telephone rescue scenario of an emergency medical dispatch center, the telephone recommendation scenario of an online sales platform, etc. For example, the online platform can obtain the target question semantics configured to guide the operator to ask the following questions by using the system provided by the present application, and the operator can determine the true purchase intention of the telephone counseling user through the questions with reference to the target question semantics.

[0031] As an example, taking the telephone rescue service of an emergency medical dispatch center as an example, the application scenario of the system in the present application that guides the operator to question the rescue sender will be described.

[0032] As shown in FIG. 1, the application scenario 100 can include a server 110, a terminal 120, and a network 130.

[0033] In multiple embodiments, the server 110 and the terminal 120 can exchange data or information via the network 130. For example, the server 110 can obtain the information and / or data in the terminal 120 via the network 130, or transmit the information and / or data to the terminal 120 via the network 130.

[0034] The terminal 120 is an electronic device configured for an operator (e.g., a dispatcher at an emergency medical dispatch center) to answer a rescue call, and can provide telephone emergency support guidance to the operator via the terminal 120. In a plurality of embodiments, the terminal 120 can obtain the operator's questions and the answers of the rescue caller, and transmit the questions and answers to the server 110 for processing. In a plurality of embodiments, the terminal 120 can present the target question semantics, the positive example probability of the person in need of rescue, the rate ratio, the rate difference, and the positive example probability trend scatter plot received from the server 110 to the operator in various ways (e.g., voice prompts, text prompts, etc.). In a plurality of embodiments, when the processing capacity of the terminal 120 is high, the terminal 120 can also process the operator's questions and the answers of the rescue caller to obtain the target question semantics, the positive example probability of the person in need of rescue, the rate ratio, the rate difference, the positive example probability trend scatter plot, etc., which is not limited to the expressions in this specification. The terminal 120 may be one of the devices with input and / or output functions, such as a mobile device, a tablet computer, etc., or any combination thereof.

[0035] The server 110 may be a single server or a server group. The server group may be centralized or distributed (e.g., the server 110 is a distributed system), may be dedicated, or may provide services simultaneously by other devices or systems. In a plurality of embodiments, the server 110 may be local or remote. In a plurality of embodiments, the server 110 may be implemented on a cloud platform or may be provided virtually. As just an example, the cloud platform may include a private cloud, a public cloud, a hybrid cloud, a community cloud, a distributed cloud, an internal cloud, a multi-layer cloud, etc., or any combination thereof.

[0036] In a plurality of embodiments, the server 110 maintains a rescue dispatch database of an emergency center and can classify patients in the dispatch database based on entity attribute clusters and entity attribute pairs. In a plurality of embodiments, the server 110 obtains a first reference group and a second reference group that need to participate in calculations from the rescue dispatch database based on questions between an operator and a rescue sender, and can transmit data information of the first reference group or the second reference group to the terminal 120.

[0037] In a plurality of embodiments, the network 130 can be any one or more of a wired network or a wireless network. For example, the network 130 can include a local area network (LAN), a wide area network (WAN), a wireless local area network (WLAN), a metropolitan area network (MAN), etc., or any combination thereof.

[0038] For ease of understanding, the technical solution of this application will be described below in conjunction with the drawings and embodiments.

[0039] FIG. 2 is an exemplary schematic diagram of a system for guiding an operator to question a rescue sender according to some embodiments of this application. As shown in FIG. 2, the system for guiding an operator to question a rescue sender includes a first reference group acquisition module 210, a classification test module 220, a receiver operating characteristic curve acquisition module 230, a candidate question semantics acquisition module 240, and a target question semantics acquisition module 250.

[0040] The first reference group acquisition module 210 is configured to select a plurality of unprocessed first reference groups from the initial reference group. The first reference group is obtained by classifying a plurality of historical rescued persons (for example, patients suspected of OHCA) in the initial reference group based on the entity attribute cluster. The entity attribute cluster is composed of a plurality of entity attribute pairs corresponding to the same core question. The core question is the subject that the operator is trying to understand through the question. For example, the core question may be a subject such as help, consciousness, breathing, heart, speech, etc. The entity attribute pair includes an entity and an attribute. The entity represents the semantics of the operator's question in the historical rescue call, and the attribute represents the semantics of the rescuer's answer to the question. For example, for the operator's question "Hello, how can I help?" and the rescuer's answer "It's very difficult. My father is not feeling well.", the semantics of "Hello" is extracted from the question, and the semantics of "not feeling well" is extracted from the answer. Therefore, the entity attribute pair corresponding to the question and the answer is "Hello - not feeling well". Since the core question of the question is "Help", the entity attribute pair corresponding to the question and the answer belongs to the "Help" entity attribute cluster. As shown in FIG. 4, the answers based on the same core question constitute an entity attribute cluster. Different entity attribute clusters are independent of each other. The coordinates (black dots) on the entity attribute cluster represent the entities extracted from the core question. Each entity corresponds to at least one different attribute, and the entity and the attribute together constitute an entity attribute pair.

[0041] During a specific implementation, for all conversations between the operator and the rescue caller, convert them into text using an algorithm from voice to text, and then perform text recognition based on the converted text to obtain questions and answers represented by the text. In multiple embodiments, the text-form questions and answers can be stored in different databases. When the rescue caller requests help by phone, call the questions and answers from the question database and the answer database respectively, and then match the questions and answers to obtain the entity attribute cluster and entity attribute pair corresponding to the conversation.

[0042] The emergency response conversations stored in the rescue dispatch database are large-scale and chaotic, making it difficult to standardize and structure them, so they are difficult to use. In the embodiments provided by this application, by structuring the emergency response conversations in the rescue dispatch database through the concepts of entity attribute clusters and entity attribute pairs, existing conversation data resources can be used to guide the commander to question the rescue caller, determine the on-site situation of the victim and the progress of the rescue of the victim by the rescue caller. As a result, the professional requirements for the emergency commander are reduced, and general commanders can provide effective emergency response services to patients.

[0043] The initial reference group is a group composed of historical rescued persons selected from the rescue dispatch database. During specific implementation, personal characteristic information of the rescued person can be extracted from the information provided by the rescue sender, and the initial reference group can be determined from a database (such as the rescue dispatch database) based on the personal characteristic information. The personal characteristic information includes, but is not limited to, information indicating personal characteristics of the rescued person such as gender, age, body type, gender, age, body type, rescue time, geographical location, etc. For example, in the case of a dialogue scenario where the operator asks "Hello, xxx command center, how can I help you?" and the rescue sender replies "It's very difficult. My father is not feeling well.", two characteristic information items of "adult" and "male" can be extracted from the semantics of the rescue sender's reply, and 500,000 cases of historical rescued persons satisfying the characteristic information are selected from the database as the initial reference group. As the dialogue progresses, a reference group more similar to the situation of the rescued person can be obtained from the database based on the entity attribute cluster and entity attribute pair corresponding to each turn of the dialogue.

[0044] The initial reference group can be further divided into a plurality of first reference groups based on the entity attribute cluster. For example, the initial reference group can include a first reference group corresponding to the "help" entity attribute cluster, a first reference group corresponding to the "consciousness" entity attribute cluster, a first reference group corresponding to the "breathing" entity attribute cluster, and so on.

[0045] During a specific implementation, a plurality of unprocessed first reference groups can be selected from an initial reference group, and then, based on the plurality of first reference groups, target question semantics can be obtained to guide an operator to ask the following questions. As just one example, the entity attribute cluster corresponding to the current dialogue is the "help" entity attribute cluster, and the first reference groups corresponding to the other four entity attribute clusters included in the initial reference group have not yet been processed. For ease of explanation, each is called a consciousness cluster, a respiration cluster, a heart cluster, and a speech cluster. Next, the first reference group corresponding to the consciousness cluster, the first reference group corresponding to the respiration cluster, the first reference group corresponding to the heart cluster, and the first reference group corresponding to the speech cluster can be selected as a plurality of first reference groups for subsequent processing.

[0046] The classification test module 220 is configured to use the positive example probability of any second reference group included in any first reference group among the plurality of first reference groups as a cut-off point, perform a classification test on any first reference group, and obtain the classification sensitivity and classification specificity corresponding to any second reference group.

[0047] As shown in FIG. 4, the second reference group is obtained by classifying the first reference group based on entity attributes. For example, the awareness cluster in the above example includes four entity attribute pairs: "coma - yes", "response - shock", "wake up - unresponsive", and "response - immobile" (the actual number of entity attribute pairs included may be much more than 4, but it is simplified for the convenience of explanation). According to the four entity attribute pairs, the first reference group corresponding to the awareness cluster can be divided into four second reference groups. Also, for example, the posture guidance cluster (entity attribute cluster) includes four entity attribute pairs: "lie down - flatten", "lie on back - it's okay now", "lie down - yes", "lie on back - OK" (the actual number of entity attribute pairs included may be much more than 4, but it is simplified for the convenience of explanation). According to the four entity attribute pairs, the first reference group corresponding to the posture guidance cluster can be divided into four second reference groups.

[0048] The positive example probability of the second reference group is the ratio of the members actually classified as positive examples to the total number of people in the second reference group, and the calculation formula is expressed as follows. P c = number of patients actually classified as positive examples / total number of patients Formula (1)

[0049] As an example, if the total number of people in a certain second reference group is 100,000 and 5,000 of its members are confirmed as positive examples, the positive example probability in this second reference group is 0.005.

[0050] Positive examples are also called "affirmative exemplifications" and are appropriate exemplifications or examples of a concept. All positive examples of each concept contain common essential features, and each concept has positive examples and negative examples (negative exemplifications). For example, elephants, lions, tigers, cats, dogs, whales, etc. are positive examples of the concept of mammals, while fish, turtles, etc. are negative examples of the concept of mammals.

[0051] In the scenario of a telephone rescue task at an emergency medical dispatch center, when a dispatcher tries to determine whether a patient is in a specific condition, the patient in that specific condition is a positive example, and the patient not in that specific condition is a negative example. For example, when a dispatcher tries to determine through conversation whether the person in need of rescue is a patient suspected of OHCA (since only emergency medical personnel can confirm whether the person in need of rescue is an OHCA patient, the dispatcher can only determine whether the person in need of rescue is a patient suspected of OHCA), positive examples are those diagnosed as OHCA patients by emergency medical personnel, and negative examples are those diagnosed as non-OHCA patients by emergency medical personnel. Also, for example, when a dispatcher tries to determine through conversation the progress of the rescue of the person in need of rescue by the rescue caller, positive examples may include patients for whom the rescue caller has started CPR, and negative examples may include patients for whom the rescue caller has not yet started CPR. Also, for example, when a dispatcher tries to determine through conversation whether the person in need of rescue has achieved ROSC, positive examples are patients who have achieved ROSC, and negative examples are patients who have not achieved ROSC.

[0052] In other application scenarios, positive or negative examples may refer to different target groups from those in the positive or negative examples in the telephone rescue task scenario of the emergency medical dispatch center, and are not limited by the expressions in this specification. For example, in the telephone recommendation scenario of an online sales platform, users with the intention to purchase a specific product are positive examples, and users without the intention to purchase a specific product are negative examples.

[0053] A preset positive example is an object assumed to be a positive example, that is, the true classification of a preset positive example may be either a positive example or a negative example. A preset negative example is an object assumed to be a negative example, that is, the true classification of a preset negative example may be either a negative example or a positive example. In the scenario of a telephone rescue task at an emergency medical dispatch center, a preset positive example is a patient who is determined and assumed by the system to be in a specific state by performing a classification test or other calculations, and a preset negative example is a patient who is determined and assumed by the system not to be in a specific state by performing a classification test or other calculations. For example, when a commander determines through dialogue whether the person in need of rescue is a patient suspected of OHCA, the preset positive example is a patient diagnosed with OHCA by a doctor determined and assumed by the system, and the preset negative example is a patient diagnosed as non-OHCA by a doctor determined and assumed by the system. Also, for example, when a commander determines through dialogue the progress of the rescue of the person in need of rescue by the rescuer, the preset positive example is a patient assumed by the system that the rescuer has started CPR, and the preset negative example is a patient assumed by the system that the rescuer has not started CPR. Also, for example, when a commander determines through dialogue whether the person in need of rescue has achieved ROSC, the preset positive example is a patient who has achieved ROSC assumed by the system, and the preset negative example is a patient who has not achieved ROSC assumed by the system.

[0054] The cut-off point is for determining the positive example label of the first reference group. In a plurality of embodiments, using the positive example probability of the second reference group as the cut-off point, members of the second reference group whose positive example probability in the first reference group is greater than or equal to the cut-off point can be used as preset positive examples, and members of the second reference group whose positive example probability in the first reference group is less than the cut-off point can be used as preset negative examples.

[0055] For example, the "Help" entity attribute cluster contains four entity attribute pairs: "Hello - fainted", "Hello - about to die", "Hello - come quickly", and "Hello - ambulance" (the actual number of entity attribute pairs included may be much more than four, but it is simplified for the convenience of explanation). After sorting, the positive example probabilities of the second reference group corresponding to each entity attribute pair are P = 0.165 for "Hello - fainted", P = 0.035 for "Hello - about to die", P = 0.019 for "Hello - come quickly", and P = 0.011 for "Hello - ambulance". When using P = 0.035 (Hello - about to die) as the cut-off point, members of the second reference group corresponding to the "Hello - fainted" entity attribute pair (whose positive example probability 0.165 is greater than the cut-off point 0.035), members of the second reference group corresponding to the "Hello - about to die" entity attribute pair (whose positive example probability 0.035 is equal to the cut-off point 0.035) can be used as preset positive examples, and members of the second reference group corresponding to the "Hello - come quickly" entity attribute pair (positive example probability 0.019 is less than the cut-off point 0.035), members of the second reference group corresponding to the "Hello - ambulance" entity attribute pair (positive example probability 0.011 is less than the cut-off point 0.035) can be used as preset negative examples.

[0056] For example, the palm positioning cluster (entity attribute cluster) contains four entity attribute pairs: "Put the wrists together and place them in the center between the patient's nipples - OK", "Place the wrists in the center between the patient's nipples - OK", "Put both hands together in front of the patient's chest - OK", and "Put both hands in front of the patient's chest - OK" (the actual number of entity attribute pairs included may be much more than 4, but it is simplified for the convenience of explanation). After sorting, the positive example probabilities of the second reference group corresponding to each entity attribute pair are P = 0.857 for "Put the wrists together and place them in the center between the patient's nipples - OK", P = 0.811 for "Place the wrists in the center between the patient's nipples - OK", P = 0.789 for "Put both hands together in front of the patient's chest - OK", and P = 0.667 for "Put both hands in front of the patient's chest - OK". When P = 0.811 (taking "Put the wrists together and place them in the center between the patient's nipples - OK" as the cut-off point), the members of the second reference group corresponding to the entity attribute pair "Put the wrists together and place them in the center between the patient's nipples - OK" (whose positive example probability is greater than the cut-off point), the members of the second reference group corresponding to the entity attribute pair "Place the wrists in the center between the patient's nipples - OK" (whose positive example probability is equal to the cut-off point) can be used as preset positive examples, and the members of the second reference group corresponding to the entity attribute pair "Place the base of the palm in front of the patient's chest - OK" (whose positive example probability is less than the cut-off point), the members of the second reference group corresponding to the entity attribute pair "Put one hand on top of the other in front of the patient's chest - OK" (whose positive example probability is less than the cut-off point) can be used as preset negative examples.

[0057] The classification test may include a screening test and a diagnostic test, and uses rapid and simple test examinations and other methods to detect unrecognized suspicious cases from seemingly healthy people. In the telephone assistance scenario of the emergency medical dispatch center, the dispatcher judges the actual condition of the assisted person through the conversation with the assistance sender. Usually, to evaluate the reliability of the classification test, it is necessary to calculate the classification sensitivity and the classification specificity. The table of the classification test is as follows.

[0058]

Table 1

[0059] During a specific implementation, the classification sensitivity can be obtained according to the number of members of the preset positive examples (true positive (a) in Table 1) actually classified as positive examples and the number of members of the first reference group actually classified as positive examples (the sum of true positive (a) and false negative (c) in Table 1). The calculation formula is as follows. TIFF2025521868000003.tif14140

[0060] During a specific implementation, the classification sensitivity can be obtained according to the number of members of the preset negative examples (true negative (d) in Table 1) actually classified as negative examples and the number of members of the first reference group (the sum of false positive (b) and true negative (d) in Table 1) actually classified as negative examples. The calculation formula is as follows. TIFF2025521868000004.tif14140

[0061] As an example, the respiratory cluster contains four entity-attribute pairs (the actual number included is much more than 4 and is simplified for convenience of explanation). After sorting based on the positive example probability of the second reference group, the positive example probabilities of each entity-attribute pair and the corresponding second reference group are "breathing-none" (P = 0.400), "breathing-uncomfortable" (P = 0.210), "wheezing-asphyxia" (P = 0.145), and "wheezing-very weak" (P = 0.074), respectively. Based on the above assumptions, using the positive example probability of the second reference group corresponding to "breathing-none" as the cut-off point, a classification test was performed on the first reference group corresponding to the "respiratory cluster", and Table 2 was obtained.

[0062]

Table 2

[0063] In Table 2, X represents the cut-off point. X = 1 means that members satisfying the cut-off point classification condition are used as preset positive examples. In this example, when taking the positive example probability (0.400) of the second reference group corresponding to "breathing - none" as the cut-off point, members of the second reference group corresponding to "breathing - none" are used as preset positive examples, and the remaining entities use members of the second reference group (positive example probability less than 0.400) corresponding to the attribute pair as preset negative examples.

[0064] In Table 2, the number of members of the preset positive examples actually classified as positive examples (a) is 480, and the number of members (a + c) of the first reference group actually classified as positive examples is 480 + 1120 = 1600. In this case, based on Equation (2), TIFF2025521868000006.tif20168 is calculated. The number of members (d) of the preset negative examples whose actual classification is a negative example is 12680. When the number of members of the first reference group actually classified as negative examples (b + d) is 720 + 12680 = 13400, based on Equation (3) TIFF2025521868000007.tif18168 can be calculated.

[0065] According to the above method, when taking the positive example probabilities of the second reference groups corresponding to "breathing - discomfort", "wheezing - asphyxia", and "wheezing - slightly weak" as the cut-off points, the classification sensitivity and classification specificity corresponding to each second reference group can be obtained.

[0066] The receiver operating characteristic curve acquisition module 230 is configured to acquire the receiver operating characteristic curve corresponding to any first reference group based on the classification sensitivity and classification specificity corresponding to a plurality of second reference groups within any first reference group.

[0067] The receiver operating characteristic curve (ROC) is a curve connecting points with the difference value between the classification specificity (Sp) obtained by a subject with different judgment criteria and 1 on the horizontal axis and the classification sensitivity (Se) on the vertical axis under specific stimulus conditions. The characteristics of the ROC curve are that it needs to pass through (0,0) and (1,1), the points on the curve are only displayed above the line connecting these two points, and it increases monotonically. Therefore, the ROC curve only makes sense when AUC > 0.5.

[0068] During specific implementation, taking the classification sensitivity of any second reference group among a plurality of second reference groups as the vertical axis and the difference value between 1 and the classification specificity of any second reference group as the horizontal axis, the coordinate points corresponding to any second reference group are obtained. Next, the coordinate points corresponding to each second reference group within the first reference group are connected to obtain the receiver operating characteristic curve corresponding to any first reference group. As just one example, using the classification sensitivity and classification specificity corresponding to a plurality of second reference groups within the first reference group corresponding to the "breathing cluster" obtained in the above example, a receiver operating characteristic curve as shown in Figure 5 can be obtained.

[0069] Based on the receiver operating characteristic curves corresponding to each group of the first reference groups within a plurality of first reference groups, the candidate question semantics acquisition module 240 determines a target first reference group from the plurality of first reference groups, and is configured such that the entities of each entity attribute pair within the entity attribute cluster corresponding to the target first reference group are used as candidate question semantics.

[0070] The larger the area under the curve of the receiver operating characteristic curve, the better the classification effect of the classification test corresponding to that curve. Therefore, in some embodiments, the area under the curve (AUC) of the receiver operating characteristic curve corresponding to each first reference group within a plurality of first reference groups is calculated (calculated using, for example, the trapezoidal rule), and then the first reference group corresponding to the maximum area under the curve can be used as the target first reference group. As a simple example, for three first reference groups corresponding to entity attribute clusters named "breathing cluster", "heart cluster", and "speech cluster" respectively, the above method is used to obtain the corresponding receiver operating characteristic curves, and the calculated AUCs of the receiver operating characteristic curves are 0.761, 0.733, and 0.662 respectively. Among them, the largest AUC corresponds to the first reference group corresponding to the "breathing cluster", so the first reference group corresponding to the "breathing cluster" is used as the target first reference group.

[0071] After the target first reference group is obtained, the entities of each entity attribute pair within the entity attribute cluster corresponding to the target first reference group can be used as candidate question semantics. As a simple example, in the "breathing cluster" corresponding to the target first reference group obtained in the above example, since it contains entity attribute pairs such as "breathing - none", "breathing - uncomfortable", "wheezing - asphyxia", and "wheezing - slightly weak", "breathing" and "wheezing" can be used as candidate question semantics. Subsequently, the target question semantics can be selected from the candidate question semantics.

[0072] The target question semantics acquisition module is configured to determine the target question semantics from the candidate question semantics based on the distance from the coordinate points corresponding to any second reference group among the plurality of second reference groups in the receiver operating characteristic curve corresponding to the target first reference group to the perfect classification coordinate points.

[0073] A perfect classification coordinate point is the coordinate point where the abscissa in the receiver operating characteristic curve is 0 and the ordinate is 1. The coordinate point (0, 1) on the receiver operating characteristic curve represents a classification test result with a classification specificity (Sp) of 1 and a classification sensitivity of 1, which is a perfect result with the highest classification accuracy. In the receiver operating characteristic curve, the distance from the coordinate point corresponding to any second reference group to the perfect classification coordinate point can reflect the accuracy of the classification result obtained by performing a classification test on the first reference group using the positive example probability of any second reference group as the cut-off point. The distance from the coordinate point corresponding to any second reference group to the perfect classification coordinate point on the receiver operating characteristic curve can be obtained by the following formula. TIFF2025521868000008.tif14138

[0074] During specific implementation, calculate the multiple distances from the coordinate points corresponding to each second reference group among the multiple second reference groups to the perfect classification coordinate point, and use the second reference group corresponding to the shortest distance among the multiple distances as the target second reference group, and the entity of the entity-attribute pair corresponding to the target second reference group can be used as the target question semantics. As a simple example, the "breathing cluster" corresponding to the target first reference group in the above example includes entity-attribute pairs such as "breathing - none", "breathing - uncomfortable", "gasping - asphyxia", "gasping - slightly weak". In the receiver operating characteristic curve, the distances from the four coordinate points corresponding to these four entity-attribute pairs to (0, 1) are 0.702, 0.548, 0.419, and 0.537 respectively. Therefore, "gasping - asphyxia" is used as the target question semantics with "gasping".

[0075] The target question semantics is configured to guide the operator to ask the following question. In multiple embodiments, the operator is the dispatcher of the emergency medical dispatch center, and the dispatcher's question is for determining whether cardiopulmonary resuscitation is necessary for the rescued person at the scene.

[0076] In multiple embodiments, the operator is the dispatcher of the emergency medical dispatch center, and the dispatcher's questions are for guiding the rescuer to perform cardiopulmonary resuscitation at the scene of the victim.

[0077] FIG. 3 is an exemplary schematic diagram of a system for guiding an operator to determine whether a victim is a positive example according to some embodiments of the present application. As shown in FIG. 3, the system for guiding an operator to determine whether a victim is a positive example includes an entity attribute cluster and entity attribute pair determination module 310, a first reference group and second reference group determination module 320, a preset positive example member number acquisition module 330, and a positive example probability determination module 340.

[0078] The entity attribute cluster and entity attribute pair determination module 310 is configured to determine the entity attribute cluster currently corresponding to the victim and the entity attribute pair currently corresponding to the victim based on the operator's questions and the rescuer's answers to those questions in any interaction between the operator and the rescuer.

[0079] The entity attribute cluster is composed of a plurality of entity attribute pairs corresponding to the same core question. The entity attribute pair includes an entity and an attribute. The entity represents the semantics of the operator's question in the historical rescue call, and the attribute represents the semantics of the rescuer's answer to the question.

[0080] During a specific implementation, the operator's questions can be obtained, the question semantic information can be extracted from the questions, the rescuer's answers to the questions can be obtained, the answer semantic information can be extracted from the answers, the core question described in the questions can be determined, the entity attribute cluster can be obtained based on the core question, and the entity attribute pair can be determined based on the question semantic information and the answer semantic information. For details of the entity attribute cluster and the entity attribute pair, refer to FIG. 2, which will not be further described here.

[0081] The first reference group and second reference group determination module 320 is configured to obtain a first reference group corresponding to an entity attribute cluster and a second reference group corresponding to an entity attribute pair from an initial reference group, where the first reference group is obtained by classifying a plurality of historical rescued persons in the initial reference group based on the entity attribute cluster, and the second reference group is obtained by classifying the first reference group based on the entity attribute.

[0082] For a detailed description of the initial reference group, the first reference group, and the second reference group, refer to FIG. 2, which will not be described again here.

[0083] The preset positive example member number acquisition module 330 is configured to acquire the number of members of the first reference group that are preset positive examples, using the positive example probability of the second reference group as a cut-off point.

[0084] The cut-off point is for determining the positive example label of the first reference group. For a detailed description of the cut-off point, refer to FIG. 2, which will not be further described here.

[0085] In a plurality of embodiments, the number of members of the second reference group in which the positive example probability in the first reference group is greater than or equal to the cut-off point can be used as the number of members of the preset positive examples. For a detailed description of this embodiment, refer to the example of step S220, which will not be described again here.

[0086] The positive example probability determination module 340 is configured to obtain the positive example probability of the rescued person based on the ratio between the number of members actually classified as positive examples among the members of the first reference group that are preset positive examples and the number of members that are preset positive examples in the first reference group.

[0087] As an example, the "Help" entity attribute cluster contains four entity attribute pairs (although the actual number of entity attribute pairs included may be much more than four, it is simplified for the convenience of explanation). After sorting based on the positive example probability of the second reference group, these four entity attribute pairs and the positive example probability of the second reference group corresponding to each entity attribute pair are respectively "Hello - fainted" (P = 0.165), "Hello - seems to be dying" (P = 0.035), "Hello - come quickly" (P = 0.019), and "Hello - ambulance" (P = 0.011). According to the current responder's answer of the rescuer "Hello - father seems to be dying", using the positive example probability (0.035) of the second reference group corresponding to the entity attribute pair "Hello - seems to be dying" as the cut-off point, the members of the second reference group corresponding to the entity attribute pairs "Hello - fainted" (the positive example probability of the corresponding second reference group is 0.165, which is greater than the cut-off point) and "Hello - seems to be dying" (the positive example probability of the corresponding second reference group is 0.035, which is equal to the cut-off point) can be used as preset positive examples, with a total of 100,000 cases. Among these 100,000 cases of members, 4,000 cases were confirmed as OHCA by doctors, and the positive example probability of the rescued person is 0.04.

[0088] The positive example probability of the rescued person is configured to guide the operator to determine whether the rescued person is a positive example. In multiple embodiments, the operator is a dispatcher of an emergency medical dispatch center, and a positive example is a rescued person who requires on-site cardiopulmonary resuscitation.

[0089] In multiple embodiments, the operator is a dispatcher of an emergency medical dispatch center, and a positive example is a patient who is encouraged by the dispatcher and for whom CPR has been initiated by the rescuer or another person on the scene.

[0090] In multiple embodiments, the operator is a dispatcher of an emergency medical dispatch center, and a positive example is a rescued person with ROSC after on-site cardiopulmonary resuscitation.

[0091] In a plurality of embodiments, a system for guiding an operator to question a rescue caller also includes a positive example probability trend scatter plot acquisition module and a slope acquisition module of a linear regression equation.

[0092] The positive example probability trend scatter plot acquisition module is configured to acquire a positive example probability trend scatter plot based on a plurality of positive example probabilities of the person in need of rescue obtained in a plurality of conversations between the operator and the rescue caller.

[0093] The slope acquisition module of the linear regression equation is configured to fit the positive example probability trend scatter plot using a linear regression equation and solve the linear regression equation to obtain the slope of the linear regression equation.

[0094] The positive example probability trend scatter plot and the slope of the linear regression equation are for representing the trend of the probability that the person in need of rescue is a positive example. During a specific implementation, coordinates on the positive example probability trend scatter plot can be obtained with the number of conversations corresponding to any one of the plurality of positive example probabilities as the abscissa and the any one of the positive example probabilities as the ordinate.

[0095] FIG. 6 shows a positive example probability trend scatter plot drawn based on a plurality of positive example probabilities of the person in need of rescue obtained in a plurality of conversations, where a, b, c, and d are positive example probability trend scatter plots obtained from a plurality of conversations between different rescue callers and the operator. Since the four slopes in FIG. 6 are 0.045, 0.0351, 0.1047, and 0.1291 respectively, based on this, it is determined that patient d has the highest possibility of getting a certain result.

[0096] Figure 7 shows another positive example probability trend scatter plot drawn based on the multiple positive example probabilities of the rescued person obtained from multiple interactions. Here, a, b, c, and d are positive example probability trend scatter plots obtained from multiple interactions between different rescue senders and operators. In Figure 7, a shows that the predicted probability of the result rapidly increases to 0.9 after only 3 iterations, indicating that the rescued patient is highly likely to belong to the actual classification result. In Figure 7, b shows that the predicted probability of the result first increases, then decreases, and then increases again, but the overall trend still tends towards the actual classification result. In Figure 7, c shows that the positive label could not be "captured" in the first few iterations, but the predicted probability of the result rapidly increased in the last iteration. In Figure 7, d shows that the possibility of belonging to the actual classification result is low. Since the slopes of the four in Figure 7 are 0.297, 0.138, 0.121, and 0.068 respectively, based on this, it is still judged that patient a has the highest possibility of becoming the result.

[0097] In the positive example probability trend scatter plot, each rescued person is independent, and the questions and answers for each entity attribute cluster are also independent. Therefore, it is assumed that the variances of the random variable P (positive example probability) at different x (abscissa) are equal. According to equations (2) and (3), by combining expertise, generally as x increases, y (y = P, 0 ≤ P ≤ 1) also increases. Therefore, the slope (b) of the linear regression equation (y = a + bx) can be used as the basis for result determination. The larger b is, the higher the positive example probability of the future rescued person will be. During specific implementation, the least squares method can be used to solve the linear regression equation and obtain the slope of the equation. The calculation formula is as follows. TIFF2025521868000009.tif19139

[0098] In multiple embodiments, other regression models can also be used to fit the scatter plot, but it is not limited to the formulas in this specification. For example, the scatter plot can be fitted using functions such as inverse functions, power functions, logarithmic functions, composite functions, growth functions, and exponential functions.

[0099] In a plurality of embodiments, the positive example probability trend scatter diagram and the slope of the linear regression equation are displayed to the operator (e.g., via the screen of the terminal 120), and the operator can be guided to determine whether the person in need of assistance is a positive example.

[0100] In a plurality of embodiments, the system for guiding the operator to question the rescue sender further includes a rate ratio acquisition module and a rate difference acquisition module.

[0101] The rate ratio acquisition module is configured to acquire a rate ratio based on the ratio of the positive example probability of the current person in need of assistance to the positive example probability of the person in need of assistance obtained in the previous interaction. The calculation formula of the rate ratio is TIFF2025521868000010.tif13139

[0102] As an example, when P1 is 0.04 and P2 is 0.1, based on Equation (6), RR can be calculated as 2.5.

[0103] The rate ratio (RR) can represent the value (contribution degree) of the cut-off point selected in this interaction for the prediction of the actual classification. When RR>1, the cut-off point selected in this interaction has discriminative value for the actual classification. When RR≦1, the cut-off point selected in this interaction has no discriminative value for the actual classification.

[0104] The rate difference acquisition module is configured to acquire a rate difference based on the difference between the positive example probability of the current person in need of assistance and the positive example probability of the person in need of assistance obtained in the previous interaction. The calculation formula of the rate difference is as follows. RD=P c -P c-1 Equation (7)

[0105] As an example, when P1 is 0.04 and P2 is 0.1, based on Equation (7), RR can be calculated as 0.06.

[0106] The rate difference (RD) can represent the value of the cut-off point selected in this conversation for predicting the actual classification. When RD > 0, the cut-off point selected in this conversation has discriminative value for the actual classification. When RR ≤ 0, the cut-off point selected in this conversation has no discriminative value for the actual classification.

[0107] In multiple embodiments, by displaying the rate difference and rate ratio to the operator, it is possible to guide the operator to judge the value for determining whether the current rescue recipient is a positive example based on the positive example probability of the current rescue recipient.

[0108] When RR > 1, the label has discriminative value for the actual classification. When RR ≤ 1, the label has no discriminative value for the actual classification. When RD > 0, the label has discriminative value for the actual classification. When RD ≤ 0, the label has no discriminative value for the actual classification.

[0109] Another embodiment of the present application provides a system for guiding an investigator to question an interviewee. The system includes a first reference group acquisition module, a classification test module, a receiver operating characteristic curve acquisition module, a candidate question semantics acquisition module, and a target question semantics acquisition module.

[0110] The first reference group acquisition module is configured to select a plurality of unprocessed first reference groups from an initial reference group.

[0111] The first reference group is obtained by classifying a plurality of historical interviewees in the initial reference group based on entity attribute clusters. The entity attribute clusters are composed of a plurality of entity attribute pairs corresponding to the same core question. The entity attribute pairs include an entity and an attribute. The entity represents the semantics of the investigator's question in a historical survey call, and the attribute represents the semantics of the interviewee's answer to the question.

[0112] For a detailed description of the initial reference group, the first reference group, and the second reference group, refer to FIG. 2, which will not be described again here.

[0113] The classification test module is configured to use the positive example probability of any second reference group included in any first reference group among the plurality of first reference groups as a cut-off point, perform a classification test on the any first reference group, and obtain the classification sensitivity and classification specificity corresponding to the any second reference group.

[0114] The cut-off point is for determining the positive example label of the first reference group, and the second reference group is obtained by classifying the first reference group based on entity attributes. For a detailed description of the second reference group and the cut-off point, refer to FIG. 2, which will not be further described here.

[0115] The receiver operating characteristic curve acquisition module is configured to obtain the receiver operating characteristic curve corresponding to the any first reference group based on the classification sensitivity and classification specificity corresponding to the plurality of second reference groups within the any first reference group.

[0116] The candidate question semantics acquisition module is configured to determine a target first reference group from the plurality of first reference groups based on the receiver operating characteristic curve corresponding to each first reference group within the plurality of first reference groups, and use the entity of each entity attribute pair within the entity attribute cluster corresponding to the target first reference group as candidate question semantics.

[0117] The target question semantics acquisition module is configured to determine target question semantics from the candidate question semantics based on the distance from the coordinate point corresponding to any second reference group among the plurality of second reference groups in the receiver operating characteristic curve corresponding to the target first reference group to the perfect classification coordinate point, where the perfect classification coordinate point is the coordinate point where the abscissa in the receiver operating characteristic curve is 0 and the ordinate is 1, and the target question semantics is configured to guide the investigator to ask the following question.

[0118] In the above embodiment of the system for guiding an investigator to ask a respondent questions, the specific processing of each module and the resulting technical effects can be referred to the relevant descriptions in the corresponding embodiment of FIG. 2, and the description is omitted.

[0119] Another embodiment of the present application provides a system for guiding an investigator to determine whether a respondent is a positive example, and the system includes an entity attribute cluster and entity attribute pair determination module, a first reference group and a second reference group determination module, a preset positive example member number acquisition module, and a positive example probability determination module.

[0120] The entity attribute cluster and entity attribute pair determination module is configured to determine the entity attribute cluster currently corresponding to the respondent and the entity attribute pair currently corresponding to the respondent based on the investigator's question and the respondent's answer to the question in any conversation between the investigator and the respondent.

[0121] The entity attribute cluster is composed of a plurality of entity attribute pairs corresponding to the same core question, and the entity attribute pair includes an entity and an attribute. The entity represents the semantics of the investigator's question in the historical survey call, and the attribute represents the semantics of the respondent's answer to the question.

[0122] For a detailed description of the entity attribute cluster and the entity attribute pair, refer to the relevant content in FIG. 2 and no further explanation will be provided here.

[0123] The first reference group and second reference group determination module is configured to obtain a first reference group corresponding to the entity attribute cluster and a second reference group corresponding to the entity attribute pair from the initial reference group.

[0124] The first reference group is obtained by classifying a plurality of historical respondents in the initial reference group based on the entity attribute cluster, and the second reference group is obtained by classifying the first reference group based on the entity attribute.

[0125] For a detailed description of the first reference group and the second reference group, refer to the relevant content in FIG. 2 and no further explanation will be provided here.

[0126] The preset positive example member number acquisition module is configured to obtain the number of members of the first reference group that are preset positive examples, using the positive example probability of the second reference group as a cut-off point.

[0127] The cut-off point is for determining the positive example label of the first reference group. For a detailed description of the cut-off point, refer to the relevant content in FIG. 2 and no further explanation will be provided here.

[0128] The positive example probability determination module is configured to obtain the positive example probability of the respondent based on the ratio between the number of members actually classified as positive examples among the members of the first reference group that are preset positive examples and the number of members of the first reference group that are preset positive examples.

[0129] The positive example probability is configured to guide the investigator to determine whether the subject is a positive example. For details of the positive example probability and positive examples, refer to the relevant content in FIG. 2 and will not be further described here.

[0130] For the specific processing of each module in the above embodiment of the system for guiding the investigator to determine whether the subject is a positive example and the technical effects thereof, reference can be made to the relevant descriptions in the corresponding embodiment in FIG. 3 and will be omitted here.

[0131] The above system for guiding the investigator to ask questions to the subject and the system for guiding the investigator to determine whether the subject is a positive example can be applied to various application scenarios, such as application scenarios of disease and health status epidemiological surveys (abbreviation of epidemiological surveys) and application scenarios of product after-sales surveys. For example, in the following, the investigation task of infectious diseases will be taken as an example for description.

[0132] Epidemiological surveys are the key to controlling infectious diseases, and the information collected from epidemiological surveys of infectious diseases can play an important role in effectively containing infectious diseases. The purpose of an epidemiological survey of infectious diseases is to clarify the movement trajectories of subjects, the people they encountered, and what happened during a certain period in the past, clarify the infection routes, identify the sources of infection, identify close contacts, take isolation measures, and provide a basis for clarifying the scope of disinfection.

[0133] In multiple embodiments, the terminal used by the investigator can obtain the investigator's questions and the respondent's answers, and send the questions and answers to the server for processing (or be processed by the terminal used by the investigator). In multiple embodiments, the terminal used by the investigator can show the investigator the target question semantics, the positive example probability of the respondent, the rate ratio, the rate difference, and the positive example probability trend scatter diagram received from the server through various methods (such as voice prompts, text prompts, etc.). This helps the investigator determine the mobile investigation situation of the respondent (for example, whether the respondent is a close contact, whether the respondent is a person who needs to take isolation measures, etc.).

[0134] In the scenario of the mobile investigation task for infectious diseases, when the investigator tries to determine whether a group of specific types of people need to take specific measures to prevent the spread of infectious diseases, the respondents belonging to this type of group are positive examples, and the respondents not belonging to this type of group are negative examples. For example, when the investigator (mobile investigator) determines through dialogue whether the respondent is a close contact, the positive example can be the respondent belonging to the close contact group, and the negative example can be the respondent belonging to the non-close contact group.

[0135] In the scenario of the mobile investigation task for infectious diseases, preset positive examples are people who are assumed to belong to a specific type of people who need to take specific measures to prevent the spread of infectious diseases for the system to perform classification tests and other calculations, and preset negative examples are people who are assumed not to belong to a specific type of people who need to take specific measures to prevent the spread of infectious diseases for the system (the terminal or server used by the investigator) to perform classification tests or other calculations. For example, when the investigator (mobile investigator) determines through dialogue whether the respondent is a close contact, the preset positive example can be the respondent belonging to the close contact assumed by the system, and the preset negative example can be the respondent belonging to the non-close contact assumed by the system.

[0136] The basic concepts have been described above, but it will be apparent to those skilled in the art that the above detailed disclosure is merely an example and is not intended to limit the present invention. Although not explicitly stated herein, those skilled in the art can make various changes, improvements, and modifications to this application. Such changes, improvements, and modifications are proposed in this application, and such changes, improvements, and modifications are still included within the spirit and scope of the exemplary embodiments of this application.

[0137] At the same time, in this application, specific words are used to describe the embodiments of this application. For example, "one embodiment", "an embodiment", and / or "a plurality of embodiments" mean specific features, structures, or characteristics related to at least one embodiment of this application. Therefore, it should be emphasized and noted that the "an embodiment" or "one embodiment" or "one alternative embodiment" mentioned multiple times in different parts of this application does not necessarily refer to the same embodiment. Also, the specific features, structures, or characteristics in one or more embodiments of this application can be appropriately combined.

[0138] Moreover, unless explicitly stated in the claims, the order of the processing elements and sequences, the use of numbers and characters, or the use of other names described in this application are not configured to limit the order of the processes and methods of this application. The above disclosure discusses some embodiments of the present invention that are considered useful currently through various examples, but such details are for the purpose of explanation only. The appended claims are not limited to the disclosed embodiments. On the contrary, it should be understood that the claims are intended to cover all modifications and equivalent combinations within the spirit and scope of the embodiments of this application. For example, the system components described above can be implemented through hardware devices, but can also be implemented through software-only solutions, such as installing the described system on an existing server or mobile device.

[0139] Similarly, in order to simplify the description disclosed in this application and thereby facilitate the understanding of one or more embodiments of the present invention, it should be noted that in the foregoing description of the embodiments of this application, multiple features may be combined in one embodiment, drawing, or its description. However, this disclosure method does not mean that the features necessary for the subject matter protected by this application are more than those mentioned in the claims. In fact, the features of the embodiments are fewer than all the features of the single embodiment disclosed above.

[0140] In some embodiments, numerical values are used to describe quantities of components and characteristics, but it should be understood that such numerical values configured to describe the embodiments are, in some examples, modified by modifiers such as "about", "approximately", or "generally". Unless otherwise specified, "about", "approximately", or "generally" means that the recited numerical value is allowed to vary by ±20%. Accordingly, in some embodiments, the numerical parameters used in the specification and claims are approximate values that may vary depending on the desired characteristics of the individual embodiments. In multiple embodiments, for numerical parameters, it is necessary to consider the specified number of significant digits and adopt a general digit-preserving method. The numerical ranges and parameters configured to confirm the breadth of the range in some embodiments of this application are approximate values, but in a particular embodiment, such numerical values are set as accurately as possible.

[0141] Each patent, patent application, patent application publication, and other materials such as articles, books, manuals, publications, documents, etc. cited in this application are hereby incorporated by reference in their entirety into this specification. Application history documents that do not match or are inconsistent with the content of this application are excluded in the same manner as documents (currently or later attached to this application) that limit the broadest scope of the claims of this application. If there is a lack of consistency or contradiction between the descriptions, definitions, and / or uses of terms in the attached materials of this application and the content described in this application, the descriptions, definitions, and / or usage methods of terms in this application shall apply.

[0142] Finally, it should be understood that the embodiments described in this application are only used to explain the principles of the embodiments of this application. Other modifications are possible within the scope of this application. Therefore, by way of example and not limitation, alternative configurations of the embodiments of this application are considered to be consistent with the teachings of this application. Therefore, the embodiments of this application are not limited to the embodiments explicitly introduced and described in this specification.

Claims

1. A system for guiding an operator to ask questions to a rescue caller, the system including a first reference group acquisition module, a classification test module, a receiver operating characteristic curve acquisition module, a candidate question semantics acquisition module, and a target question semantics acquisition module, the first reference group acquisition module is configured to select a plurality of unprocessed first reference groups from an initial reference group, the first reference group being obtained by classifying a plurality of historical rescued persons in the initial reference group based on entity attribute clusters, the entity attribute clusters being composed of a plurality of entity attribute pairs corresponding to the same core question, the entity attribute pairs including an entity and an attribute, the entity representing the semantics of an operator's question in a historical rescue call, and the attribute representing the semantics of a rescue caller's answer to the question, the classification test module is configured to perform a classification test on any one of the plurality of first reference groups by using the positive example probability of any second reference group included in any one of the plurality of first reference groups as a cut-off point to obtain the classification sensitivity and classification specificity corresponding to the any second reference group, the cut-off point being for determining the positive example label of the first reference group, and the second reference group being obtained by classifying the first reference group based on entity attributes, the receiver operating characteristic curve acquisition module is configured to obtain a receiver operating characteristic curve corresponding to any one of the plurality of first reference groups based on the classification sensitivity and classification specificity corresponding to a plurality of second reference groups within the any one of the plurality of first reference groups, the candidate question semantics acquisition module is configured to determine a target first reference group from the plurality of first reference groups based on the receiver operating characteristic curve corresponding to the first reference group within the plurality of first reference groups, and use the entity of each entity attribute pair within the entity attribute cluster corresponding to the target first reference group as candidate question semantics, The target question semantics acquisition module is configured to determine target question semantics from the candidate question semantics based on the distance from the coordinate point corresponding to any second reference group among the plurality of second reference groups in the receiver operating characteristic curve corresponding to the target first reference group to the perfect classification coordinate point, where the perfect classification coordinate point is a coordinate point with an abscissa of 0 and an ordinate of 1 in the receiver operating characteristic curve, and the target question semantics is configured to guide the operator to ask the next question. A system characterized by this is provided.

2. Taking the positive example probability of any second reference group included in any first reference group among the plurality of first reference groups as a cut-off point, performing a classification test on the any first reference group to obtain the classification sensitivity and classification specificity corresponding to the any second reference group, Regarding members of the second reference group whose positive example probability in the first reference group is greater than or equal to the cut-off point as preset positive examples, and regarding members of the second reference group whose positive example probability in the first reference group is less than the cut-off point as preset negative examples, Obtaining the classification sensitivity according to the number of members of the preset positive examples actually classified as positive examples and the number of members of the first reference group actually classified as positive examples, Obtaining the classification specificity according to the number of members of the preset negative examples actually classified as negative examples and the number of members of the first reference group actually classified as negative examples. The system according to claim 1 is characterized by including these.

3. Obtaining the receiver operating characteristic curve corresponding to any first reference group based on the classification sensitivity and classification specificity corresponding to the plurality of second reference groups within the any first reference group, Taking the classification sensitivity of any second reference group among the plurality of second reference groups as the ordinate, taking the absolute value of the difference between the classification specificity of the any second reference group and 1 as the abscissa, and obtaining the coordinate point corresponding to the any second reference group. Connecting the coordinate points corresponding to each second reference group within the first reference group to obtain the receiver operating characteristic curve corresponding to the arbitrary first reference group, The system according to claim 2, characterized in that it includes this.

4. Based on the receiver operating characteristic curves corresponding to each first reference group within the plurality of first reference groups, determining a target first reference group from the plurality of first reference groups, Calculating the area under the curve of the receiver operating characteristic curve corresponding to each first reference group within the plurality of first reference groups, Using the first reference group corresponding to the maximum area under the curve among the areas under the curves as the target first reference group, The system according to claim 1, characterized in that it includes this.

5. Based on the distance from the coordinate point corresponding to an arbitrary second reference group among the plurality of second reference groups in the receiver operating characteristic curve corresponding to the target first reference group to the perfect classification coordinate point, determining the target question semantics from the candidate question semantics, Calculating a plurality of distances from the coordinate point corresponding to each second reference group among the plurality of second reference groups to the perfect classification coordinate point, Using the second reference group corresponding to the shortest distance among the plurality of distances as the target second reference group, and using the entity of the entity-attribute pair corresponding to the target second reference group as the target question semantics, The system according to claim 1, characterized in that it includes this.

6. Extracting the personal characteristic information of the rescued person from the information spoken by the rescuer, and determining the initial reference group based on the personal characteristic information, thereby obtaining the initial reference group, The system according to claim 1, characterized in that it is like this.

7. A system for guiding an operator to determine whether the rescued person is a positive example, the system includes an entity-attribute cluster and an entity-attribute pair determination module, a first reference group and a second reference group determination module, a preset positive example member number acquisition module and a positive example probability determination module, The entity attribute cluster and entity attribute pair determination module are configured to determine an entity attribute cluster currently corresponding to the victim and an entity attribute pair currently corresponding to the victim based on a question of the operator and an answer of the rescuer to the question in any interaction between the operator and the rescuer. Here, the entity attribute cluster is composed of a plurality of entity attribute pairs corresponding to the same core question. The entity attribute pair includes an entity and an attribute. The entity represents the semantics of the operator's question in the historical rescue call, and the attribute represents the semantics of the rescuer's answer to the question. The first reference group and second reference group determination module is configured to obtain a first reference group corresponding to the entity attribute cluster and a second reference group corresponding to the entity attribute pair from an initial reference group. Here, the first reference group is obtained by classifying a plurality of historical victims in the initial reference group based on the entity attribute cluster, and the second reference group is obtained by classifying the first reference group based on the entity attribute. The preset positive example member number acquisition module is configured to acquire the number of members of the first reference group that are preset positive examples, with the positive example probability of the second reference group as a cut-off point. Here, the cut-off point is for determining the positive example label of the first reference group. The positive example probability determination module is configured to obtain the positive example probability of the victim based on the ratio between the number of members actually classified as positive examples among the members of the first reference group that are preset positive examples and the number of members that are preset positive examples in the first reference group. The positive example probability is configured to guide the operator to determine whether the victim is a positive example. A system characterized by this.

8. The system further includes a positive example probability trend scatter diagram acquisition module and a slope acquisition module for a linear regression equation. The positive example probability trend scatter plot acquisition module is configured to acquire a positive example probability trend scatter plot based on a plurality of positive example probabilities of the person in need of rescue obtained in a plurality of conversations between the operator and the rescue sender. The slope acquisition module of the linear regression equation is configured to fit the positive example probability trend scatter plot using a linear regression equation and solve the linear regression equation to obtain the slope of the linear regression equation. Here, the positive example probability trend scatter plot and the slope of the linear regression equation are for representing the trend of the probability that the person in need of rescue is a positive example. The system according to claim 7 is characterized in that.

9. Acquiring a positive example probability trend scatter plot based on a plurality of positive example probabilities of the person in need of rescue obtained in a plurality of conversations between the operator and the rescue sender is The system according to claim 8, characterized in that it includes obtaining coordinates on the positive example probability trend scatter plot with the number of conversations corresponding to an arbitrary positive example probability among the plurality of positive example probabilities as the abscissa and the arbitrary positive example probability as the ordinate.

10. Determining the entity attribute cluster currently corresponding to the person in need of rescue and the entity attribute pair currently corresponding to the person in need of rescue is Obtaining the operator's question and extracting question semantics information from the question, Obtaining the rescue sender's answer to the question and extracting answer semantics information from the answer, Determining the core question described in the question and obtaining the entity attribute cluster based on the core question, The system according to claim 7, characterized in that it includes determining the entity attribute pair based on the question semantics information and the answer semantics information.

11. Taking the positive example probability of the second reference group as a cut-off point and obtaining the number of members of the first reference group who are preset positive examples is The system according to claim 7, characterized in that it includes taking the number of members of the second reference group whose positive example probability in the first reference group is greater than or equal to the cut-off point as the number of members of the preset positive example.

12. The system further includes a rate ratio acquisition module and a rate difference acquisition module. The ratio acquisition module is configured to acquire a ratio based on the ratio of the current positive example probability of the assisted person to the positive example probability of the assisted person obtained in the previous conversation. The difference acquisition module is configured to acquire a difference according to the difference between the current positive example probability of the assisted person and the positive example probability of the assisted person obtained in the previous conversation. The system according to claim 7, wherein the difference and the ratio are configured to guide the operator to determine the value of whether the current assisted person is a positive example based on the current positive example probability of the assisted person.

13. A system for guiding an investigator to ask questions to an investigated person, the system including a first reference group acquisition module, a classification test module, a receiver operating characteristic curve acquisition module, a candidate question semantics acquisition module, and a target question semantics acquisition module. The first reference group acquisition module is configured to select a plurality of unprocessed first reference groups from an initial reference group. Here, the first reference group is obtained by classifying a plurality of historical investigated persons in the initial reference group based on an entity attribute cluster. The entity attribute cluster is composed of a plurality of entity attribute pairs corresponding to the same core question. The entity attribute pair includes an entity and an attribute. The entity represents the semantics of the investigator's question in a historical survey call, and the attribute represents the semantics of the investigated person's answer to the question. The target question semantics acquisition module is configured to determine target question semantics from the candidate question semantics based on the distance from the coordinate point corresponding to any second reference group among the plurality of second reference groups in the receiver operating characteristic curve corresponding to the target first reference group to a perfect classification coordinate point. Here, the perfect classification coordinate point is a coordinate point where the abscissa in the receiver operating characteristic curve is 0 and the ordinate is 1. The target question semantics is configured to guide the investigator to ask the next question.

14. A system for guiding an investigator to determine whether an interviewee is a positive example, the system including an entity attribute cluster and entity attribute pair determination module, a first reference group and second reference group determination module, a preset positive example member number acquisition module, and a positive example probability determination module. The entity attribute cluster and entity attribute pair determination module is configured to determine an entity attribute cluster currently corresponding to the interviewee and an entity attribute pair currently corresponding to the interviewee based on the investigator's questions and the interviewee's answers to those questions in any interaction between the investigator and the interviewee. Here, the entity attribute cluster is composed of a plurality of entity attribute pairs corresponding to the same core question. The entity attribute pair includes an entity and an attribute. The entity represents the semantics of the investigator's questions in a historical survey call, and the attribute represents the semantics of the interviewee's answers to the questions. The first reference group and second reference group determination module is configured to obtain a first reference group corresponding to the entity attribute cluster and a second reference group corresponding to the entity attribute pair from an initial reference group. Here, the first reference group is obtained by classifying a plurality of historical interviewees in the initial reference group based on the entity attribute cluster, and the second reference group is obtained by classifying the first reference group based on the entity attribute. The positive example probability determination module is configured to obtain the positive example probability of the interviewee based on the ratio between the number of members actually classified as positive examples among the members of the first reference group that are preset positive examples and the number of members that are preset positive examples in the first reference group. The positive example probability is configured to guide the investigator to determine whether the interviewee is a positive example. A system characterized by this.

Images