Railway production scheduling incoming call intelligent answering recognition device and method

Abstract

The application relates to the technical field of call answering identification, in particular to a device and method for intelligently identifying railway production dispatch incoming calls, which specifically comprises the following steps: obtaining voice information of a current railway production dispatch incoming call and information records of each previous railway production dispatch incoming call, obtaining complexity of the current dispatch incoming call voice information based on occurrence of each keyword in the current dispatch task and occurrence frequency confusion of all keywords in the current dispatch task, determining the number of events used for searching based on the complexity of the current dispatch incoming call voice information, so as to determine more suitable reference historical dispatch records for the current dispatch task, and the dispatcher can refer to the records, thereby avoiding the problem that the final search result cannot better serve as reference for the dispatcher due to the fact that the fine standard of detection cannot be determined in the traditional way, and the dispatch response efficiency is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of call answering recognition, and in particular to a device and method for intelligent call answering recognition applied to railway production dispatching. BACKGROUND

[0002] The device and method for intelligent call answering recognition applied to railway production dispatching is an important aspect of the intelligent and automated development of the railway industry. With the growth of railway transportation demand and the increasing complexity of operational management, traditional dispatching methods have been unable to meet the efficient and precise dispatching needs. Therefore, the introduction of intelligent call answering recognition technology has become an inevitable trend in the industry. Currently, the railway industry is actively exploring and applying various intelligent technologies to improve dispatching efficiency and safety. For example, through voice recognition technology, the dispatcher's voice instructions are converted into text information, and then through natural language processing technology, these information is analyzed and processed to realize real-time monitoring of train operation status and automatic execution of dispatching instructions.

[0003] Currently, the answering of railway production dispatching calls is usually handled by dispatchers responding to the problems reflected by the dispatching calls according to the rules and processes determined by railway production. Considering the differences in proficiency and experience of dispatchers in current dispatching work, they may not be able to respond quickly, which will affect the efficiency of railway production dispatching response. At the same time, dispatchers usually refer to similar dispatching records to extract helpful dispatching processing schemes or suggestions. However, current methods cannot determine appropriate screening criteria based on the current dispatching call and provide corresponding appropriate reference schemes for dispatchers, which affects the efficiency of production dispatching response. SUMMARY

[0004] To solve the above technical problems, the purpose of the present application is to provide a device and method for intelligent call answering recognition applied to railway production dispatching, which adopts the following technical solutions:

[0005] In a first aspect, the present application provides a method for intelligent call answering recognition applied to railway production dispatching, which includes the following steps:

[0006] Obtain the voice information of the current railway production dispatching call and the information record of each historical railway production dispatching call, denoted as current dispatching call voice information and historical dispatching call information record, respectively, and obtain the keywords;

[0007] Denote the text content of the current dispatching call voice information as the current dispatching task; determine the event occurrence frequency of the current dispatching task based on the occurrence of each keyword in the current dispatching task;

[0008] determine the event irrelevance degree of the current scheduling task based on the occurrence of each keyword in the history of each time of scheduling the incoming call information record in the current scheduling task; determine the event low-frequency consistency degree of the current scheduling task based on the confusion of the occurrence frequency of all keywords in the current scheduling task; and determine the complexity of the current scheduling task based on the event occurrence frequency, the event irrelevance degree and the event low-frequency consistency degree of the current scheduling task.

[0009] take a preset length of time each day as a period; calculate the peak period sensitivity of each period in a day according to the number of railway production scheduling calls in each period in a day; determine the number of subsequent event retrievals of the current scheduling task based on the complexity of the current scheduling task, the number of keywords and the peak period sensitivity;

[0010] determine the keywords for subsequent retrieval in the current scheduling task based on the number of subsequent event retrievals.

[0011] In one embodiment, the process of obtaining keywords is as follows:

[0012] take the railway production scheduling knowledge base as the input of the keyword extraction algorithm, and take each keyword in the scheduling knowledge base as the output; for any word in the scheduling problem of the current scheduling call voice information and the history of each time of scheduling the incoming call information record, if the any word is the same as any keyword in the scheduling knowledge base, take the any word as the keyword in the scheduling problem of the current scheduling call voice information or the history of each time of scheduling the incoming call information record.

[0013] In one embodiment, the event occurrence frequency is the average of the occurrence frequencies of all keywords in the current scheduling task.

[0014] In one embodiment, the process of obtaining the event irrelevance degree is as follows:

[0015] for a keyword i in the current scheduling task, obtain the number of common occurrences of each keyword and the keyword i in all history of scheduling the incoming call information record, denoted as the first absolute number; take the maximum value of all the first absolute numbers of the keyword i as the absolute occurrence number of the keyword i.

[0016] the process of obtaining the relative occurrence number of each keyword in the current scheduling task is as follows:

[0017] P1, randomly obtain a first keyword from the current scheduling task, denoted as the first keyword;

[0018] P2, divide the first keyword into the determined keyword;

[0019] P3, take all the keywords in the current scheduling task except the determined keyword as the undetermined keyword.

[0020] P4, obtaining the number of common occurrences of each uncertain keyword and the first keyword in all historical scheduling incoming call information records, denoted as a first relative number; obtaining the maximum value of all the first relative numbers of the first keyword as the relative occurrence number of the first keyword; and denoting the uncertain keyword corresponding to the maximum value as a second keyword;

[0021] P5, adding the second keyword into the determined keyword, and repeating steps P3-P5 until each keyword in the current scheduling task obtains a corresponding relative occurrence number, and then stopping;

[0022] denoting the event irrelevancy degree of the current scheduling task as W, and the expression of W being: wherein I represents the number of keywords in the current scheduling task, w' i is the relative occurrence number of the i-th keyword of the current scheduling task, w i is the absolute occurrence number of the i-th keyword of the current scheduling task.

[0023] In one embodiment, the expression of the event low-frequency consistency degree is:

[0024] E = 1 - norm(σ q ), wherein E represents the event low-frequency consistency degree of the current scheduling task, norm() is a normalization function, and σ q is the standard deviation of the occurrence frequency of all keywords in the current scheduling task.

[0025] In one embodiment, the expression of the complexity of the current scheduling task is:

[0026] R = Q x W + (1 - Q) x E, wherein R represents the complexity of the current scheduling task, Q represents the event occurrence frequency of the current scheduling task, W represents the event irrelevancy degree of the current scheduling task, and E represents the event low-frequency consistency degree of the current scheduling task.

[0027] In one embodiment, the process of obtaining the sensitivity during the peak period of each period in a day is:

[0028] calculating the ratio of the number of railway production scheduling incoming calls in each period of each day to the total number of railway production scheduling incoming calls in the day as the peak proportion of each period of each day in the railway production scheduling incoming calls of the day; and denoting the sensitivity during the peak period of the k-th period in a day as U k , and the expression of U k is:

[0029] wherein norm() is a normalization function, P represents a preset number of days, x p,kThe peak proportion of the railway production scheduling call of the kth period on the pth day.

[0030] In one embodiment, the expression of the number of subsequent event searches of the current scheduling task is:

[0031] In the formula, J represents the number of subsequent event searches of the current scheduling task, I represents the number of keywords in the current scheduling task, R represents the complexity of the current scheduling task, U' represents the sensitivity during the peak period of the period in which the current time is located, represents the upward rounding symbol, and norm() is a normalization function.

[0032] In one embodiment, the determination process of the keywords for subsequent searches in the current scheduling task is:

[0033] The number of subsequent event searches of the current scheduling task is recorded as the first number, all keywords in the current scheduling task are arranged in a keyword sequence in descending order of the event occurrence frequency of the keywords, and the first number of keywords in the keyword sequence are used as the keywords for subsequent searches.

[0034] In a second aspect, the embodiments of the present application further provide a railway production scheduling call intelligent answering recognition device, which comprises a memory, a processor, and a computer program stored in the memory and running on the processor, and the processor implements the steps of the method according to any one of the above embodiments when executing the computer program.

[0035] The embodiments of the present application have at least the following beneficial effects:

[0036] The present application obtains the voice information of the current railway production scheduling call and the information record of each historical railway production scheduling call, determines the complexity of the current scheduling call voice information based on the occurrence of each keyword in the current scheduling task in the information record of each historical scheduling call, and the confusion of the occurrence frequency of all keywords in the current scheduling task, considers the complexity of the scheduling task under the conditions that the overall event occurrence frequency is high and low, improves the rationality of the keyword search number setting when the event complexity is different, determines the number of events used for searching based on the complexity of the current scheduling call voice information, and determines a more adaptive reference historical scheduling record for the current scheduling task to provide a reference for the dispatcher, avoids the problem that the final search result cannot better serve as a reference for the dispatcher due to the inability to determine the fine standard in the traditional method, and improves the scheduling response efficiency. BRIEF DESCRIPTION OF DRAWINGS

[0037] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor.

[0038] Figure 1 The step flow chart of the method for intelligently answering and identifying incoming calls applied to railway production scheduling is provided for an embodiment of the present application.

[0039] Figure 2 The schematic diagram for determining the keywords for subsequent retrieval. DETAILED DESCRIPTION

[0040] In order to further illustrate the technical means and effects adopted by the present application to achieve the predetermined purpose of the application, the following will combine the drawings and the preferred embodiments to specifically describe the application of the device and method for intelligently answering and identifying incoming calls applied to railway production scheduling, the specific implementation, structure, features and effects thereof, as follows. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0041] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present application belongs.

[0042] The specific scheme of the device and method for intelligently answering and identifying incoming calls applied to railway production scheduling provided by the present application will be specifically described below in combination with the drawings.

[0043] Please refer to Figure 1 , which shows the step flow chart of the method for intelligently answering and identifying incoming calls applied to railway production scheduling provided by an embodiment of the present application. The method comprises the following steps:

[0044] In step S1, the voice information of the current time railway production scheduling incoming call and the information record of each historical railway production scheduling incoming call are obtained, which are respectively recorded as the current scheduling incoming call voice information and the historical each scheduling incoming call information record, and the keywords are obtained respectively.

[0045] If the current time dispatcher receives the railway production scheduling incoming call, the voice information in the incoming call is obtained, which is recorded as the current scheduling incoming call voice information, and the voice information is converted into text form by the voice recognition technology of Keda Xunfei, so as to carry out subsequent analysis and processing. The voice recognition technology of Keda Xunfei is a known technology, and the specific process will not be described again.

[0046] It should be noted that for the recognition of the voice, the present application only provides a voice recognition method, there are many existing voice recognition methods, and the implementer can also use other voice recognition techniques for voice text conversion, which is not specifically limited by the present application.

[0047] Meanwhile, when the railway production dispatches the incoming call, the dispatcher will record the content of the incoming call, therefore, the information in each railway production dispatch incoming call in the historical time is obtained according to the dispatcher's incoming call record, which is recorded as the historical each dispatch incoming call information record, wherein the dispatch problem and the corresponding solution are contained in each recorded incoming call information.

[0048] The text content corresponding to the current dispatch incoming call voice information and the dispatch problem in the historical each dispatch incoming call information record are respectively processed by JIEBA word segmentation technology; the railway production dispatch knowledge base is the prior knowledge about the railway production dispatch, and the railway production dispatch knowledge base is taken as the input of the TF-IDF algorithm for keyword extraction, and the output is the keyword contained in the dispatch knowledge base; if each word in the text content of the current dispatch incoming call voice information is the same as any keyword in the dispatch knowledge base, then the word is the keyword in the text content of the current dispatch incoming call voice information; similarly, if each word in the dispatch problem of the historical each dispatch incoming call information record is the same as any keyword in the dispatch knowledge base, then the word is the keyword in the dispatch problem of the historical each dispatch incoming call information record. Wherein, the JIEBA word segmentation technology and the TF-IDF algorithm are both known technologies, and the specific process will not be repeated.

[0049] Step S2, the text content of the current dispatch incoming call voice information is recorded as the current dispatch task; the event occurrence frequency of the current dispatch task is determined based on the occurrence of each keyword in the current dispatch task.

[0050] For the railway production dispatch, it involves the staff of each production system site of the railway, which will report and reflect various types of problems through the incoming call mode, such as passengers, freight, passenger transport, etc., so many types of problems will bring certain dispatch pressure to the dispatcher. In order to improve the work efficiency of the dispatcher, the current railway incoming call voice information is matched with the appropriate historical dispatch incoming call information record, so as to provide certain reference and help for the dispatcher, and further improve the dispatch response speed.

[0051] Since the keywords contained in the task content reported in the dispatch call can determine the type of task to be handled in the dispatch call, such as train operation, cargo loading and unloading, maintenance, etc., each task has its specific requirements and process, and the number of keywords in the dispatch call determines the complexity of the dependency relationship between the dispatch tasks, such as the delay of a train may affect the operation plan of other trains, and means that the availability and distribution of resources such as track, station, locomotive, personnel, etc. need to consider these dependency relationships to develop a dispatch plan. The occurrence frequency of the keywords can reflect the historical comparison cases of the corresponding similar cases, therefore, based on the keywords in the text content of the current dispatch call voice information, the event occurrence frequency of the current dispatch call voice information is determined, the method is:

[0052] The text content of the current dispatch call voice information is recorded as the current dispatch task; the occurrence frequency of each keyword in the current dispatch task is obtained, and the expression of the event occurrence frequency of the current dispatch task is:

[0053] In the formula, Q represents the event occurrence frequency of the current dispatch task, I represents the number of keywords in the current dispatch task, q i represents the occurrence frequency of the i-th keyword in the current dispatch task.

[0054] Step S3, determining the event irrelevance degree of the current dispatch task based on the occurrence of each keyword in the historical dispatch call information record of the current dispatch task; constructing the event low-frequency consistency degree of the current dispatch task based on the confusion of the occurrence frequency of all keywords of the current dispatch task; constructing the complexity of the current dispatch task based on the event occurrence frequency, the event irrelevance degree and the event low-frequency consistency degree of the current dispatch task.

[0055] A keyword corresponds to an event, and in daily railway dispatching work, some events often have relevance, that is, some events often occur together in dispatching work, and therefore the keywords corresponding to these events also often occur together in the dispatch call information. Therefore, when there are events that do not often occur together in the dispatch task, it means that it may correspond to a more complex and difficult dispatch task.

[0056] To analyze the mutual influence relationship between each event, first, the absolute occurrence times of each keyword in the current dispatch task are obtained, which are as follows:

[0057] Taking keyword i in the current dispatch task as an example, the common occurrence times of each keyword and keyword i in all historical dispatch call information records are obtained, which are recorded as the first absolute times; the maximum value of all the first absolute times of keyword i is taken as the absolute occurrence times of keyword i.

[0058] wherein the co-occurrence refers to if the keyword i appears in a historical scheduled incoming call information record, each keyword other than the keyword i in the historical scheduled incoming call information record is a co-occurrence keyword of the keyword i;

[0059] The co-occurrence frequency refers to if the keyword i and a certain keyword co-occur in X historical scheduled incoming call information records, the co-occurrence frequency of the keyword i and the certain keyword is X.

[0060] Then, the relative occurrence frequency of each keyword in the current scheduling task is obtained, specifically:

[0061] P1, a first keyword is randomly obtained from the current scheduling task, and is recorded as a first keyword;

[0062] P2, the first keyword is classified into the determined keyword;

[0063] P3, all the keywords other than the determined keyword in the current scheduling task are obtained as the undetermined keyword;

[0064] P4, the co-occurrence frequency of each undetermined keyword and the first keyword in all historical scheduled incoming call information records is obtained, which is recorded as a first relative frequency; the maximum value of all the first relative frequencies of the first keyword is obtained as the relative occurrence frequency of the first keyword; the undetermined keyword corresponding to the maximum value is recorded as a second keyword;

[0065] P5, the second keyword is classified into the determined keyword, and steps P3-P5 are repeated until the relative occurrence frequency of each keyword in the current scheduling task is obtained.

[0066] Wherein, for the last keyword in the current scheduling task, if the relative occurrence frequency cannot be obtained by the above method, the co-occurrence frequency of the last keyword and each keyword in the current scheduling task is obtained, and the maximum value is obtained as the relative occurrence frequency of the last keyword.

[0067] Finally, the expression of the event irrelevance degree of the current scheduling task is:

[0068] In the formula, W represents the event irrelevance degree of the current scheduling task, I represents the number of keywords in the current scheduling task, w' i is the relative occurrence frequency of the i th keyword of the current scheduling task, w i is the absolute occurrence frequency of the i th keyword of the current scheduling task.

[0069] Further, the event low-frequency consistency degree of the current scheduling task is calculated:

[0070] In addition to the above analysis of the degree of irrelevance of the time of the current scheduling task, the low-frequency consistency of the current scheduling task should also be considered, which means that when the events contained in the current scheduling task all maintain a very low frequency, it indicates that the current scheduling task may face a relatively rare situation, and lacks the support and assistance of historical reference information. By analyzing the low-frequency consistency, the overall low frequency of the current scheduling task caused by the low frequency of a certain event can be avoided. The expression of the low-frequency consistency of the events of the current scheduling task is:

[0071] E=1-norm(σ q ), wherein E represents the low-frequency consistency of the events of the current scheduling task, norm() is a normalization function, and σ q is the standard deviation of the frequency of occurrence of all keywords in the current scheduling task.

[0072] Further, the complexity of the current scheduling task is calculated:

[0073] Through the above process, the complexity of the current scheduling task can be determined. Based on the frequency of occurrence of events in the current scheduling task, when the frequency of occurrence of events is high, the scheduling task complexity under high frequency should be analyzed from the perspective of simultaneous occurrence of common related events. The higher the frequency of occurrence of events, the more attention should be paid to the degree of irrelevance of the events of the current scheduling task. When the frequency of occurrence of events is low, the scheduling task complexity should be analyzed from the perspective of rare occurrence of events. The lower the frequency of occurrence of events, the less reference information is available, and the more attention should be paid to the low-frequency consistency of the events of the current scheduling task. Accordingly, the complexity of the current scheduling task is constructed, and the expression is:

[0074] R=Q×W+(1-Q)×E, wherein R represents the complexity of the current scheduling task, Q represents the frequency of occurrence of events of the current scheduling task, W represents the degree of irrelevance of events of the current scheduling task, and E represents the low-frequency consistency of events of the current scheduling task.

[0075] Step S4: Taking a preset length of time each day as a period; calculating the peak period sensitivity of each period in a day according to the number of calls of railway production scheduling in each period of the day; determining the number of subsequent event retrievals of the current scheduling task based on the complexity of the current scheduling task, the number of keywords, and the peak period sensitivity.

[0076] In the railway production scheduling task, the railway production operation is different in different time periods, and thus the problems are different. In the peak period, continuous production operation of the railway can cause production docking, coordination and other events. Although the railway production operation can be different every day, the overall operation mode is relatively fixed. Therefore, the number of incoming calls for production scheduling in each period in the near future can be used to define the peak period of each day.

[0077] Firstly, the last week is analyzed, and each half hour is taken as a period length, for example, 0.00-0.30 of each day is the first period, and 0.31-1.00 is the second period. The number of incoming calls for railway production scheduling in each period of each day is obtained.

[0078] It should be noted that the number of days and the length of the period can be set by the implementer according to the actual situation in other embodiments of the present application, and the present application does not make specific limitations.

[0079] Further, the ratio of the number of incoming calls for railway production scheduling in each period of each day to the total number of incoming calls for railway production scheduling in the day is calculated as the peak proportion of each period of each day in the incoming calls for railway production scheduling in the day.

[0080] Further, the expression of the peak period sensitivity of each period in a day is:

[0081]

[0082] In the formula, U k represents the peak period sensitivity of the kth period in a day; norm() is a normalization function; P represents a preset number of days, and the value of P in the embodiment of the present application is 7; x p,k represents the peak proportion of the kth period of the pth day in the incoming calls for railway production scheduling, and the peak period sensitivity of the kth period in different days is the same.

[0083] For each period of railway production scheduling, the higher the peak proportion of the incoming calls for production scheduling in each day in the near future, the more the period is in the peak period of railway production scheduling, that is, the higher the peak period sensitivity.

[0084] Then, the complexity of the current scheduling task can be determined through the above process, and based on this, the fineness of the railway production scheduling screening can be adjusted. The higher the complexity of the current scheduling task and the higher the fineness of the screening should be ensured when the current scheduling task is in the peak period, so as to ensure that the adaptability of the historical scheduling records provided to the dispatcher is higher, and the work efficiency of the dispatcher is improved.

[0085] The expression for determining the number of subsequent event retrievals based on the complexity of the current scheduling task is:

[0086] In the formula, J represents the number of subsequent event searches after the current scheduling task, I represents the number of keywords in the current scheduling task, R represents the complexity of the current scheduling task, U' represents the sensitivity during the peak period of the period in which the current time is located, represents the upward rounding symbol, and norm() is a normalization function.

[0087] The number of subsequent event searches in the current scheduling task is thus determined, and the fine degree of different screening can be met.

[0088] In step S5, the number of keywords for subsequent search in the current scheduling task is determined based on the number of subsequent event searches.

[0089] All keywords in the current scheduling task are arranged in a keyword sequence in descending order of event occurrence frequency of the keywords, and the first J keywords in the keyword sequence are used as the keywords for subsequent search.

[0090] The screening is performed according to the high and low of the event occurrence frequency in the current scheduling task, and the higher the event occurrence frequency, the more the keyword should be used for search, because it usually corresponds to more historical reference cases, and is more helpful for the dispatcher to make reference and comparison. Furthermore, different screening fine degrees are provided to provide different degrees of similar scheduling tasks for the dispatcher to make reference, so that the number and accuracy of historical scheduling records provided to the dispatcher can be balanced.

[0091] A determination process of the keywords for subsequent search is shown in FIG. 1. Figure 2

[0092] Based on the same inventive concept as the above method, the embodiments of the present application also provide a device for intelligent answering recognition of incoming calls in railway production scheduling, which comprises a memory, a processor, and a computer program stored in the memory and running on the processor, and the processor implements the steps of any one of the methods in the above method for intelligent answering recognition of incoming calls in railway production scheduling when executing the computer program.

[0093] ​To sum up, the embodiment of the application provides an intelligent answering call recognition method applied to railway production scheduling, acquires voice information of a current railway production scheduling call and information records of historical railway production scheduling calls each time, acquires complexity of current scheduling call voice information based on occurrence of each keyword in the current scheduling task in the information record of the historical scheduling call each time and confusion of occurrence frequencies of all keywords of the current scheduling task, considers complexity of the scheduling task under two conditions that the overall occurrence frequency of the event is high and low, improves rationality of setting the number of keyword retrievals when the complexity of the event is different, determines the number of events used for retrieval based on complexity of the current scheduling call voice information, and thus determines more adaptive reference historical scheduling records for the current scheduling task, so as to provide reference for the dispatcher, avoids the problem that the final retrieval result cannot better provide reference for the dispatcher due to the fact that the fine standard of detection cannot be determined in the traditional way, and improves scheduling response efficiency.

[0094] It should be noted that the above-mentioned sequence of the embodiments of the application is only for description, and does not represent the advantages and disadvantages of the embodiments. The above describes the specific embodiments of the application. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multi-task processing and parallel processing are possible or can be advantageous.

[0095] Each of the embodiments in the application is described in a progressive manner, and the same and similar parts of each of the embodiments can be referred to each other. Each of the embodiments mainly describes the difference from other embodiments.

[0096] The above only describes the preferred embodiments of the application, and does not limit the application. Any modification, equivalent replacement, improvement, etc. made within the principles of the application shall be included in the protection scope of the application.

Claims

1. A method for identifying incoming calls intelligently for railway production scheduling, characterized in that, The method comprises the following steps: Acquire the voice information of the current railway production dispatch call and the information record of each historical railway production dispatch call, and mark them as current dispatch call voice information and historical dispatch call information record respectively, and acquire the keywords respectively; Mark the text content of the current dispatch call voice information as the current dispatch task; determine the event occurrence frequency of the current dispatch task based on the occurrence of each keyword in the current dispatch task; determine the event irrelevance degree of the current scheduling task based on the occurrence of each keyword in the history of each scheduling of the incoming call information record; calculate the event low-frequency consistency degree of the current scheduling task, expressed as: , in the formula, E represents the event low-frequency consistency degree of the current scheduling task, is a normalization function, is the standard deviation of the occurrence frequency of all keywords in the current scheduling task; calculate the complexity of the current scheduling task, expressed as: , in the formula, R represents the complexity of the current scheduling task, Q represents the event occurrence frequency of the current scheduling task, W represents the event irrelevance degree of the current scheduling task, and E represents the event low-frequency consistency degree of the current scheduling task. Take the preset time length of each day as a period; The ratio of the number of times of railway production scheduling calls in each period of a day to the total number of times of railway production scheduling calls in the day is calculated as the peak proportion of railway production scheduling calls in the day for each period of the day; the sensitivity during the peak period of the kth period in a day is recorded as , The expression of the sensitivity is: , wherein, is a normalization function, P represents a preset number of days, represents a peak proportion of the railway production dispatch call of the kth period on the pth day; the number of subsequent event searches after the current dispatch task is calculated, and the expression is: , wherein, J represents the number of subsequent event searches after the current dispatch task, I represents the number of keywords in the current dispatch task, and R represents the complexity of the current dispatch task, represents the sensitivity during the peak period of the period in which the current time is located, represents a rounding up symbol, is a normalization function; Determine the keywords for subsequent retrieval in the current dispatch task based on the number of subsequent event retrievals.

2. The method for identifying the incoming call intelligent receiving of the application to the railway production scheduling according to claim 1, wherein, The process of acquiring the keywords is as follows: Take the railway production dispatch knowledge base as the input of the keyword extraction algorithm, and the output is each keyword in the dispatch knowledge base; for any word in the dispatch problem of the current dispatch call voice information and the historical dispatch call information record, if the any word is the same as any keyword in the dispatch knowledge base, the any word is taken as the keyword in the dispatch problem corresponding to the current dispatch call voice information or the historical dispatch call information record.

3. The method for intelligent answering call identification applied to railway production dispatching call according to claim 1, characterized in that, The event occurrence frequency is the average of the occurrence frequencies of all keywords in the current dispatch task.

4. The method for intelligent answering call identification applied to railway production dispatching call according to claim 1, characterized in that, The process of acquiring the event irrelevance degree is as follows: For a keyword i in the current dispatch task, acquire the common occurrence times of each keyword and the keyword i in all historical dispatch call information records, and mark them as the first absolute times; take the maximum value of all the first absolute times of the keyword i as the absolute occurrence times of the keyword i. The process of acquiring the relative occurrence times of each keyword in the current dispatch task is as follows: P1, randomly acquire the first keyword from the current dispatch task, and mark it as the first keyword; P2, classify the first keyword into the determined keywords; P3, acquire all the keywords in the current dispatch task except the determined keywords as the undetermined keywords; P4, acquire the common occurrence times of each undetermined keyword and the first keyword in all historical dispatch call information records, and mark them as the first relative times; acquire the maximum value of all the first relative times of the first keyword as the relative occurrence times of the first keyword; mark the undetermined keyword corresponding to the maximum value as the second keyword; P5, classify the second keyword into the determined keywords, and repeat steps P3-P5 until each keyword in the current dispatch task acquires the corresponding relative occurrence times; The degree of event irrelevance of the current scheduling task is denoted as W, and the expression of W is as follows: , wherein I represents the number of keywords in the current scheduling task, is the relative occurrence frequency of the i th keyword of the current scheduling task, is the absolute occurrence frequency of the i th keyword of the current scheduling task.

5. The method for intelligent answering call identification applied to railway production dispatching call according to claim 1, characterized in that, The process of determining the keywords for subsequent retrieval in the current dispatch task is as follows: Mark the number of subsequent event retrievals of the current dispatch task as the first number, arrange all the keywords in the current dispatch task in descending order of the event occurrence frequency of the keywords to form a keyword sequence, and take the first number of keywords in the keyword sequence as the keywords for subsequent retrieval.

6. The application is applied to the intelligent answering recognition device of the incoming call of the railway production scheduling, including memory, processor and computer program stored in the memory and running on the processor, characterized in that, The processor executes the computer program to realize the steps of the method in any one of claims 1-5.

Images