CN116596274A - Police dispatch method, police dispatch equipment and storage medium - Google Patents
Police dispatch method, police dispatch equipment and storage medium Download PDFInfo
- Publication number
- CN116596274A CN116596274A CN202310700886.XA CN202310700886A CN116596274A CN 116596274 A CN116596274 A CN 116596274A CN 202310700886 A CN202310700886 A CN 202310700886A CN 116596274 A CN116596274 A CN 116596274A
- Authority
- CN
- China
- Prior art keywords
- information
- dispatch
- police
- alarm
- case
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 48
- 238000001514 detection method Methods 0.000 claims abstract description 20
- 230000001105 regulatory effect Effects 0.000 claims abstract description 4
- 239000013598 vector Substances 0.000 claims description 39
- 230000008451 emotion Effects 0.000 claims description 17
- 238000013527 convolutional neural network Methods 0.000 claims description 12
- 239000011159 matrix material Substances 0.000 claims description 12
- 238000012545 processing Methods 0.000 claims description 12
- 230000007246 mechanism Effects 0.000 claims description 5
- 239000000523 sample Substances 0.000 claims 1
- 238000012549 training Methods 0.000 description 12
- 230000008569 process Effects 0.000 description 7
- 230000003993 interaction Effects 0.000 description 6
- 238000010586 diagram Methods 0.000 description 4
- 238000000605 extraction Methods 0.000 description 4
- 206010063659 Aversion Diseases 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 230000004927 fusion Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 238000002360 preparation method Methods 0.000 description 2
- 239000002699 waste material Substances 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000007635 classification algorithm Methods 0.000 description 1
- 238000013145 classification model Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 230000007935 neutral effect Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 238000011176 pooling Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06311—Scheduling, planning or task assignment for a person or group
- G06Q10/063112—Skill-based matching of a person or a group to a task
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/22—Procedures used during a speech recognition process, e.g. man-machine dialogue
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Educational Administration (AREA)
- Physics & Mathematics (AREA)
- Economics (AREA)
- Tourism & Hospitality (AREA)
- Entrepreneurship & Innovation (AREA)
- Health & Medical Sciences (AREA)
- Marketing (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Development Economics (AREA)
- General Business, Economics & Management (AREA)
- Game Theory and Decision Science (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Computational Linguistics (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Human Computer Interaction (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to a police dispatch method, police dispatch equipment and a storage medium. The method comprises the following steps: responding to the accessed alarm telephone, and identifying the voice information based on the voice intelligent identification model to obtain an identification result; the identification result comprises semantic information and alarm association information; determining case information according to the semantic information and the alarm associated information; wherein the case information comprises a case type; acquiring police personnel information of a corresponding district according to the case information, and matching the police personnel information with the case type to generate pre-dispatch information; the police information includes tampering with the regulatory regulations; and obtaining the manual dispatch information corresponding to the alarm telephone and the pre-dispatch information to perform similarity detection, and determining an alarm instruction according to a similarity detection result. The technical scheme of the invention can send pre-dispatch information for the alarm to prepare the district which is likely to give out the alarm, distribute police officers and accurately give out the alarm according to the case type.
Description
Technical Field
The present invention relates to the field of data processing technologies, and in particular, to a police service scheduling method, apparatus, and storage medium.
Background
The current police dispatch method is: the receiving personnel receives the alarm telephone of the principal to obtain the case information of the alarm case, then selects the idle police personnel from the surrounding of the case occurrence place, sends information to the police personnel through the police terminal to inform the case occurrence place and the contact way of the principal, so that the case information reaches the case occurrence place to process the alarm case.
The receiving personnel need to listen to the voice information of the alarming personnel and judge the alarming event based on the heard voice, and issue the alarming event task to the district. The jurisdiction can only receive police information issued or transferred by the alarm center, and the police information generally comprises: the location and the event cannot be determined, and the alarm is difficult to be given out in a targeted manner.
Disclosure of Invention
The invention provides a police dispatch method, police dispatch equipment and a storage medium, and aims to improve the dispatching speed and give an alarm for a case in dispatching.
In a first aspect, an embodiment of the present invention provides a police service scheduling method, including:
responding to the accessed alarm telephone, and identifying the voice information based on the voice intelligent identification model to obtain an identification result; the identification result comprises semantic information and alarm identity association information;
determining case information according to the semantic information and the alarm identity association information; wherein the case information comprises a case type;
acquiring police personnel information of a corresponding district according to the case information, and matching the police personnel information with the case type to generate pre-dispatch information; the police information includes tampering with the regulatory regulations;
and obtaining the manual dispatch information corresponding to the alarm telephone and the pre-dispatch information to perform similarity detection, and determining an alarm instruction according to a similarity detection result.
In a second aspect, an embodiment of the present invention provides an electronic device, including:
one or more processors;
a memory for storing one or more programs;
the one or more programs, when executed by the one or more processors, cause the one or more processors to implement the police dispatch method as provided by any embodiment of the present invention.
In a third aspect, embodiments of the present invention provide a storage medium containing computer executable instructions which, when executed by a computer processor, are used to perform a police dispatch method as provided by any of the embodiments of the present invention.
According to the police dispatch method, the police dispatch equipment and the storage medium, the alarm voice is identified through the voice intelligent identification model, the case type is determined, the pre-dispatch is carried out, the alarm instruction is determined, the problems of delay in alarm giving and poor pertinence are solved, the pre-dispatch information is realized to prepare jurisdictions in which the police possibly can be given, the alarm giving speed is improved, and the police is accurately given according to the case type.
Drawings
FIG. 1 is a flowchart of a police dispatch method according to a first embodiment of the present invention;
fig. 2 is a schematic structural diagram of a police service system according to a first embodiment of the present invention;
FIG. 3 is a flowchart of a registration phase of a voice intelligent recognition model according to a first embodiment of the present invention;
FIG. 4 is a flowchart of a recognition stage of a voice intelligent recognition model according to a first embodiment of the present invention;
fig. 5 is a schematic structural diagram of a similarity detection model according to a first embodiment of the present invention;
fig. 6 is a schematic structural diagram of an electronic device according to a second embodiment of the present invention.
Detailed Description
The invention is described in further detail below with reference to the drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and are not limiting thereof. It should be further noted that, for convenience of description, only some, but not all of the structures related to the present invention are shown in the drawings.
Example 1
Fig. 1 is a flowchart of a police service scheduling method provided in a first embodiment of the present invention, where the method may be applicable to a case of scheduling police service by a police service system, and the method may be performed by a computer device, for example, a server, as shown in fig. 2, where the server includes a cloud server and a plurality of jurisdiction servers, the cloud server may receive an alarm call, the jurisdiction servers communicate with terminals of the police service, and the police service system may be arranged in a different manner, and may not be provided with the cloud server, but synchronize voice information to each jurisdiction server, and perform voice recognition by the jurisdiction servers to determine whether the case is managed by the jurisdiction server. The method specifically comprises the following steps:
step 110, responding to the accessed alarm telephone, and identifying the voice information based on the voice intelligent identification model to obtain an identification result;
the identification result comprises semantic information and alarm identity association information. By identifying the voice information, the semantic information comprises event content, event places, event characters and the like, and the information associated with the alarm can be known through the voice information, wherein the information comprises identity, gender, emotion and the like. In addition, the voice signal often contains abundant environmental background sounds, such as various noises, sounds of non-alarms, and the like. The voice intelligent recognition model can remove noise of the alarm telephone and recognize semantic information and information related to the identity of the alarm.
Step 120, determining case information according to the semantic information and the alarm associated information;
wherein the case information includes a case type. The district server may have a classification library corresponding to the case types. Different case types are set according to preset regulations.
Step 130, acquiring police personnel information of a corresponding district according to the case information, and matching the police personnel information with the case type to generate pre-dispatch list information;
wherein the police information includes a tamper-evident code. The district server can set case labels for police officers, and the district personnel can modify the police officer police labels of different police officers through the district server, so that the police officers can be preferentially dispatched with case types good at the police officers. And constructing the corresponding relation between the case-adept label and the case type classification library. Therefore, the overlapping police officers can be selected through the correspondence between the case label and the case type classification library, and the overlapping police officers can be determined to be possible police officers to be served as targets of pre-dispatch. It should be noted that, the pre-dispatch list is to inform the police officer meeting the alarm condition that the police officer needs to make preparation or send the pre-dispatch list to the route leading to the alarm place, and send the pre-dispatch list information to the corresponding district server, screen the task information of the police officer through the district server, and send the pre-dispatch list information to the mobile terminal associated with the police officer meeting the alarm condition, where the mobile terminal may be a mobile phone downloading police APP or a police terminal corresponding to the police officer system.
The recognition result includes semantic information and alarm related information, the semantic information includes time content, time and place, event tasks, the alarm related information includes (alarm emotion, sex), some recognition items may have situations where recognition is unclear, for example, there may be two places recognized by a voice recognition model: and when the pre-dispatch is sent, the pre-dispatch is sent to the district A and the district B simultaneously.
The pre-dispatch information may be in the following several cases: if the alerter is a female and the emotion has any emotion of liveliness, fear of sadness and aversion, at least one female police officer is required to send an alarm in the pre-dispatch information so as to calm the emotion of the alerter; if the distance between the recognized case occurrence place and the plurality of jurisdictions is smaller than the preset distance (for example, 3 km), and the event task is any one of a large case and a plurality of cases, sending pre-dispatch information to all jurisdictions with the case occurrence place smaller than the preset distance; when emotion is happy or boring, the pre-dispatch information is temporarily not dispatched, the problem that alarming personnel waste police force for false alarming is avoided, and the information is dispatched manually.
And 140, obtaining the manual dispatch information corresponding to the alarm telephone and the pre-dispatch information to perform similarity detection, and determining an alarm instruction according to a similarity detection result.
After receiving the alarm call, the receiver generates manual dispatch information based on the voice information received by the receiver. And the receiver generates manual dispatch information according to the received voice information and sends the manual dispatch information to a cloud server corresponding to the police service system. Comparing the pre-dispatching information with the manual dispatching information in the cloud server, issuing an alarm instruction to the district server, and forwarding the alarm instruction to a terminal of police officers by the district server.
According to the technical scheme, when the receiver answers the alarm call, the voice information is identified, and the pre-dispatch information is generated and sent to the corresponding district, so that police officers in the district can make police preparation. And the similarity detection is carried out by adopting manual dispatch and pre-dispatch information, the final dispatch is determined, and the police dispatch is pointed out to avoid omission of police caused by simply relying on intelligent dispatch. The server utilizes the advantage of operation to identify the voice information, shortens the identification time and improves the identification speed and accuracy.
Optionally, the identifying the voice information based on the voice intelligent identification model to obtain an identification result includes:
inputting the voice information into a semantic recognition model to obtain semantic information, wherein the semantic information comprises event content, event places and event characters;
inputting the voice information into an identity association recognition model to obtain the alarm identity association information, wherein,
inputting the voice information into a first SVM classifier for gender matching to obtain a corresponding gender label;
and respectively inputting the voice information into a second SVM classifier and a third SVM classifier which correspond to the gender label for emotion matching and speaker matching, and voting to determine a classification result by using a minimum distance criterion to obtain a corresponding emotion label and identity label.
The intelligent speech recognition model adopted by one implementation manner in the embodiment can be a baseline system, and the preset times of training are performed by adopting a speech training set, wherein the speech training set comprises neutral, angry, fear, happiness, boring, sadness and aversion emotion. In the training process, three categories of acoustic features of gender, emotion and identity are combined in series to form a fusion feature vector for multidimensional information recognition, and fig. 3 and fig. 4 are a baseline system adopted by a voice intelligent recognition model, wherein the input of the baseline system is the fusion feature vector, and the system is a brand new system constructed by connecting a plurality of SVM classifiers in series and parallel and comprises three types of speaker gender, emotion and identity recognition. As shown in fig. 3, in the registration stage, for SVM training, gender classification is performed by using the combination features of multidimensional speaker information recognition, a classification model is stored, then, according to the classification result, all the voice samples with the same gender are integrated together to form two voice library subsets marked with male and female labels, then, classification of speaker identity and emotion is performed in redefined male and female corpus data respectively, and finally, the model training result based on gender is sequentially stored as a model 2 and a model 3. Fig. 4 is a sample recognition stage, at this time, the voice to be tested is firstly subjected to gender matching according to the model 1 to obtain the tag 1 corresponding to the gender, then is subjected to emotion and speaker matching respectively according to the two models stored in the training stage, and the classification result is determined by voting according to the minimum distance criterion, so as to finally obtain the corresponding emotion tag 2 and identity tag 3. In addition, the multi-dimensional judgment is different from the traditional single-dimensional judgment, the results of three labels are required to be considered simultaneously, and the average value is calculated to obtain the final judgment result.
The baseline system is essentially an algorithm for solving the multi-label classification, and applies the thought of the problem conversion method. First, multi-dimensional speaker information recognition is considered as a classification problem for three kinds of tags. Then, through the processing process of voice waveform data such as proper feature extraction, the original multi-dimensional information classification problem is converted into the problem of utilizing a plurality of multi-classification algorithms which are existing at present and are developed more mature. The trained intelligent recognition model can accurately recognize the gender, emotion and identity of the alarm.
Optionally, the determining case information according to the semantic information and the alarm association information includes:
determining event content and the case type according to the event content, the event location, the event person and the alarm associated information; wherein the case type is set according to preset regulation regulations.
Optionally, the obtaining police personnel information of a corresponding district according to the case information and matching with the case type, and generating the pre-dispatch information includes:
determining a corresponding district according to the event location;
acquiring police personnel information of a corresponding district from a police personnel database;
comparing the rule and regulation treaty of police officers with the case type, and determining the police officers as police officers.
The police officers who are good at processing legal regulations and overlap with the case types are determined to be police officers, and the police officers who are good at processing the case types process the events, so that the system is more targeted and beneficial to the efficiency and quality of case processing.
Optionally, the police personnel information further comprises a working state, a working time and an alert age time;
after comparing the policeman's tampering with the legal regulations with the case type, determining the policeman, the method further comprises:
if the working states of the police officers are all police outputting states, calculating case matching degree in police officers which do not output police in the corresponding jurisdiction, and selecting police officers with highest case matching degree as the police officers, wherein the case type matching degree, the working time length matching degree and the police age time length matching degree are respectively summed with the corresponding matching weight products to obtain the case matching degree.
If overlapping police officers are in an alarm state, and no processing personnel corresponding to the case types exist at the moment, calculating case matching degree among the rest people without alarm, and selecting the person with the highest case matching degree to alarm, wherein the case matching degree comprises case type matching degree, working time length matching degree and alarm time length matching degree, and each matching degree corresponds to a matching weight, namely case matching degree = case type matching degree, 1+ working time length matching degree x matching weight 2+ alarm time length matching degree x matching weight 3.
Optionally, according to the preset rule corresponding to the case type, judging the number difference between the rule numbers and the rule numbers which are good for police officers to process, and determining the matching degree of the case type;
determining the matching degree of the working time length according to the matching degree between the average processing time length of the case type and the residual working time length of police officers;
and determining that the matching degree of the alert age duration is higher according to the requirement of the case importance degree on the alert age duration.
And judging the criminal law number or public security management rule number of the possible offence according to the voice recognition result, and judging the numerical difference between the criminal law number and the police officer's code number of the legal regulations, wherein the case type matching degree is higher when the numerical difference is smaller. The working time length matching degree is the matching degree between the average processing time length of the case of the type and the rest working time length of the police officer in big data statistics, and if the rest working time length is longer than the average processing time length and the time length difference is smaller, the working time length matching degree is higher, so that the waste of police strength can be avoided, and the police officer is prevented from overtime for a long time to influence the normal rest of the police officer. The remaining working time of the police officer can be determined according to the current time and the shift of the police officer, for example, the shift of the police officer is from eight early to five afternoon, the current time is four afternoon, and the remaining working time is 1 hour. The matching degree of the police age duration is related to the importance degree of the case, and burst class or a plurality of people need police service personnel with longer police age duration to process. It should be noted that, the longer the alert age time is, the higher the matching degree of the corresponding alert age time is, for example, the shorter the alert age is, the better the physical strength of the corresponding police officer is, and the higher the matching degree of the alert age time is. The matching weights 1, 2 and 3 can be correspondingly modified according to different characteristics of each district so as to adapt to the characteristics of the cases in the district. For example, most cases managed in a certain district are detected cases, the matching weight 3 can be properly adjusted to be high, so that more experienced police officers can rapidly give an alarm, for example, the former time of cases in a certain district is relatively large in quantity, and the police officers are relatively high in physical effort, and the matching weight 2 can be properly increased to meet the rest requirements of the police officers.
Optionally, the obtaining the manual dispatch information corresponding to the alarm phone and the pre-dispatch information to perform similarity detection, and determining an alarm instruction according to a similarity detection result includes:
if the similarity between the manual dispatch information and the pre-dispatch information is greater than a first preset threshold, the pre-dispatch information is considered as a final dispatch, and the warning instruction is issued to the district corresponding to the final dispatch;
if the similarity between the manual dispatch information and the pre-dispatch information is smaller than the first preset threshold and larger than the second preset threshold, comparing the similarity information in the event location information set, and sending the alarm instruction to a district where the event location in the manual dispatch information coincides with the event location in the pre-dispatch information;
and if the similarity between the manual dispatch information and the pre-dispatch information is smaller than the second preset threshold, the manual dispatch information is considered as a final dispatch, and the warning instruction is issued to the district corresponding to the final dispatch.
Optionally, the obtaining the manual dispatch information corresponding to the alarm phone and the pre-dispatch information to perform similarity detection includes:
generating a first text vector corresponding to the pre-form information and a second text vector corresponding to the manual form information based on a word vector model;
extracting first local feature information of the first text vector and second local feature information of the second text vector by using a preset convolutional neural network;
extracting first self-attention information of the first text vector and second self-attention information of the second text vector by adopting a self-attention mechanism;
splicing the first text vector, the first local feature information and the first self-attention information to obtain a first semantic matrix, and splicing the second text vector, the second local feature information and the second self-attention information to obtain a second semantic matrix;
extracting first characteristic information and second characteristic information from the first semantic matrix and the second semantic matrix by adopting a two-way long-short-term memory network;
extracting first semantic features and second semantic features from the first feature information and the second feature information, connecting the first semantic features and the second semantic features to obtain feature vectors, transmitting the feature vectors into a classifier to classify, and calculating cosine phase speeds between texts.
The method comprises the steps of adopting a model shown in fig. 5 for information similarity detection, generating Word vectors by pre-dispatch information and manual dispatch information through a Word2Vec model, then transmitting the Word vectors into an information interaction model for training, adopting PSO-CNN for optimization training by an information interaction layer, adding information between the attention mechanism and the self-attention mechanism to increase statement information, adopting BiLSTM extraction features after information combination, finally carrying out pooling, and adopting cosine similarity calculation and prediction to obtain the matching degree of two texts.
The vector generation layer of the model is used for representing the text into a word vector form, so that the semantic expression of the subsequent vector is facilitated. Converting text into vectors may employ Word2Vec methods or Glove pre-training methods, etc. The Word2Vec method is mainly adopted in the model in the embodiment, and is simple and efficient, and the Word vector representation with higher precision can be obtained from the corpus. The method adopts a sliding window mode to conduct feature extraction, and two training modes exist, namely, a word of a context is predicted according to a target word, and the other word of the context is predicted according to the word of the context.
In the information interaction layer, the local characteristic information of the extracted text adopts a convolutional neural network, and for short text, the characteristic of the local information can sometimes represent the whole sentence meaning of the text, so that the model also extracts the local characteristic of the text to carry out information interaction aiming at the semantic diversity problem in text matching. Because the structure of the convolutional neural network is complex, and the structure of the convolutional neural network is determined by the super-parameters, the super-parameters cannot be obtained by training, and are generally set by relying on the existing experience. Different tasks are performed on different data sets, and the set hyper-parameters are also different. Therefore, before experiments are carried out, the convolutional neural network is adopted to match the text matching data set used in the text matching data set, the particle swarm optimization algorithm is adopted to combine with the convolutional neural network model to optimize the learning rate, and the learning rate of the convolutional neural network suitable for text matching on the data set is obtained after a plurality of iterations. Then, the local feature information 1 and the local feature 2 of the text vector 1 and the text vector 2 are extracted by using a convolutional neural network (PSO-CNN), and the self-attention information 1 and the self-attention information 2 of the text vector 1 and the text vector 2 are extracted by adopting a self-attention mechanism to calculate the features of the semantic information of a deeper level. The text vector 1, the local feature information 1 and the self-attention information are spliced to obtain a semantic matrix 1, the text vector 2, the local feature information 2 and the self-attention information are spliced to obtain a semantic matrix 2, and the semantic matrix 1 and the semantic matrix 2 adopt BiLSTM to extract the feature information 1 and the feature information 2.
The interaction and prediction layer mainly connects the extracted semantic features and then calculates the text matching result. The semantic features 1 and 2 are extracted from the feature information 1 and the feature information 2 by adopting an average_mapping method, the two semantic features are connected to obtain feature vectors, the feature vectors are transmitted into a classifier to be classified, the cosine phase speed between texts is calculated, and then the information similarity of the manual dispatch information can be obtained.
In the similarity detection model adopted in the embodiment, the order information is subjected to vectorization processing, and the two order information are matched with the semantic information of the lower level of the other party. And then extracting structural feature information of the text by using PSO-CNN. And in order to capture long-distance dependence of the text, the word order and the context information between the texts are considered in the matching process, and in order to obtain deeper semantic information, interactions are performed inside the text. In order to solve the problem of text structure, bi LSTM is used for feature collection and extraction, and the problem of long-distance dependence of partial sentences is solved. The similarity of the order information can be judged more accurately according to the phrases and the contexts, and the obtained result is more accurate.
Example two
Fig. 6 is a schematic structural diagram of an electronic device according to a second embodiment of the present invention, as shown in fig. 6, the electronic device includes a processor 610, a memory 620, an input device 630 and an output device 640; the number of processors 610 in the electronic device may be one or more, one processor 610 being taken as an example in fig. 6; the processor 610, memory 620, input device 630, and output device 640 in the electronic device may be connected by a bus or other means, for example in fig. 6.
The memory 620 is a computer readable storage medium, and may be used to store software programs, computer executable programs, and modules, such as program instructions/modules corresponding to the police dispatch method in the embodiments of the present invention. The processor 610 executes various functional applications of the electronic device and data processing, i.e., implements the police dispatch method described above, by running software programs, instructions, and modules stored in the memory 620.
Memory 620 may include primarily a program storage area and a data storage area, wherein the program storage area may store an operating system, at least one application program required for functionality; the storage data area may store data created according to the use of the terminal, etc. In addition, memory 620 may include high-speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other non-volatile solid-state storage device. In some examples, memory 620 may further include memory remotely located relative to processor 610, which may be connected to the electronic device via a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
The input device 630 may be used to receive input numeric or character information and to generate key signal inputs related to user settings and function control of the electronic device. The output device 640 may include a display device such as a display screen.
Example III
A third embodiment of the present invention also provides a storage medium containing computer-executable instructions, which when executed by a computer processor, are for performing a police dispatch method comprising:
responding to the accessed alarm telephone, and identifying the voice information based on the voice intelligent identification model to obtain an identification result; the identification result comprises semantic information and alarm identity association information;
determining case information according to the semantic information and the alarm identity association information; wherein the case information comprises a case type;
acquiring police personnel information of a corresponding district according to the case information, and matching the police personnel information with the case type to generate pre-dispatch information; the police information includes tampering with the regulatory regulations;
and obtaining the manual dispatch information corresponding to the alarm telephone and the pre-dispatch information to perform similarity detection, and determining an alarm instruction according to a similarity detection result.
Of course, the storage medium containing the computer executable instructions provided in the embodiments of the present invention is not limited to the method operations described above, and may also perform the related operations in the police dispatch method provided in any embodiment of the present invention.
From the above description of embodiments, it will be clear to a person skilled in the art that the present invention may be implemented by means of software and necessary general purpose hardware, but of course also by means of hardware, although in many cases the former is a preferred embodiment. Based on such understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art in the form of a software product, which may be stored in a computer readable storage medium, such as a floppy disk, a Read-Only Memory (ROM), a random access Memory (Random Access Memory, RAM), a FLASH Memory (FLASH), a hard disk or an optical disk of a computer, etc., and include several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method according to the embodiments of the present invention.
While the invention has been described in detail in the foregoing general description, embodiments and experiments, it will be apparent to those skilled in the art that modifications and improvements can be made thereto. Accordingly, such modifications or improvements may be made without departing from the spirit of the invention and are intended to be within the scope of the invention as claimed.
Claims (10)
1. A police dispatch method comprising:
responding to the accessed alarm telephone, and identifying the voice information based on the voice intelligent identification model to obtain an identification result; the identification result comprises semantic information and alarm identity association information;
determining case information according to the semantic information and the alarm identity association information; wherein the case information comprises a case type;
acquiring police personnel information of a corresponding district according to the case information, and matching the police personnel information with the case type to generate pre-dispatch information; the police information includes tampering with the regulatory regulations;
and obtaining the manual dispatch information corresponding to the alarm telephone and the pre-dispatch information to perform similarity detection, and determining an alarm instruction according to a similarity detection result.
2. The method of claim 1, wherein the identifying the voice information based on the voice intelligent identification model to obtain the identification result comprises:
inputting the voice information into a semantic recognition model to obtain semantic information, wherein the semantic information comprises event content, event places and event characters;
inputting the voice information into an identity association recognition model to obtain the alarm identity association information, wherein,
inputting the voice information into a first SVM classifier for gender matching to obtain a corresponding gender label;
and respectively inputting the voice information into a second SVM classifier and a third SVM classifier which correspond to the gender label for emotion matching and speaker matching, and voting to determine a classification result by using a minimum distance criterion to obtain a corresponding emotion label and identity label.
3. The method of claim 2, wherein the determining case information based on the semantic information and the alarm associated information comprises:
determining event content and the case type according to the event content, the event location, the event person and the alarm associated information; wherein the case type is set according to preset regulation regulations.
4. A method according to claim 2 or 3, wherein the obtaining police information corresponding to the district and matching with the case type according to the case information, and generating pre-dispatch information comprises:
determining a corresponding district according to the event location;
acquiring police personnel information of a corresponding district from a police personnel database;
comparing the rule and regulation treaty of police officers with the case type, and determining the police officers as police officers.
5. The method of claim 4, wherein the police information further comprises an operational status, an operational time period, and an alert age time period;
after comparing the policeman's tampering with the legal regulations with the case type, determining the policeman, the method further comprises:
if the working states of the police officers are all police outputting states, calculating case matching degree in police officers which do not output police in the corresponding jurisdiction, and selecting police officers with highest case matching degree as the police officers, wherein the case type matching degree, the working time length matching degree and the police age time length matching degree are respectively summed with the corresponding matching weight products to obtain the case matching degree.
6. The method of claim 5, wherein the step of determining the position of the probe is performed,
judging the number difference between the rule code numbers of the polices and the rule code numbers of the polices according to the preset rule code corresponding to the case type, and determining the matching degree of the case type;
determining the matching degree of the working time length according to the matching degree between the average processing time length of the case type and the residual working time length of police officers;
and determining that the matching degree of the alert age duration is higher according to the requirement of the case importance degree on the alert age duration.
7. The method of claim 1, wherein the obtaining the manual dispatch information corresponding to the alert phone and the pre-dispatch information for similarity detection, and determining the alert command according to the similarity detection result, comprises:
if the similarity between the manual dispatch information and the pre-dispatch information is greater than a first preset threshold, the pre-dispatch information is considered as a final dispatch, and the warning instruction is issued to the district corresponding to the final dispatch;
if the similarity between the manual dispatch information and the pre-dispatch information is smaller than the first preset threshold and larger than the second preset threshold, comparing the similarity information in the event location information set, and sending the alarm instruction to a district where the event location in the manual dispatch information coincides with the event location in the pre-dispatch information;
and if the similarity between the manual dispatch information and the pre-dispatch information is smaller than the second preset threshold, the manual dispatch information is considered as a final dispatch, and the warning instruction is issued to the district corresponding to the final dispatch.
8. The method of claim 7, wherein the obtaining the manual dispatch information corresponding to the alert phone and the pre-dispatch information for similarity detection comprises:
generating a first text vector corresponding to the pre-form information and a second text vector corresponding to the manual form information based on a word vector model;
extracting first local feature information of the first text vector and second local feature information of the second text vector by using a preset convolutional neural network;
extracting first self-attention information of the first text vector and second self-attention information of the second text vector by adopting a self-attention mechanism;
splicing the first text vector, the first local feature information and the first self-attention information to obtain a first semantic matrix, and splicing the second text vector, the second local feature information and the second self-attention information to obtain a second semantic matrix;
extracting first characteristic information and second characteristic information from the first semantic matrix and the second semantic matrix by adopting a two-way long-short-term memory network;
extracting first semantic features and second semantic features from the first feature information and the second feature information, connecting the first semantic features and the second semantic features to obtain feature vectors, transmitting the feature vectors into a classifier to classify, and calculating cosine phase speeds between texts.
9. An electronic device, comprising:
one or more processors;
a memory for storing one or more programs;
the one or more programs, when executed by the one or more processors, cause the one or more processors to implement the police dispatch method of any one of claims 1-8.
10. A storage medium containing computer executable instructions which, when executed by a computer processor, are for performing the police dispatch method of any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310700886.XA CN116596274A (en) | 2023-06-14 | 2023-06-14 | Police dispatch method, police dispatch equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310700886.XA CN116596274A (en) | 2023-06-14 | 2023-06-14 | Police dispatch method, police dispatch equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116596274A true CN116596274A (en) | 2023-08-15 |
Family
ID=87600908
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310700886.XA Pending CN116596274A (en) | 2023-06-14 | 2023-06-14 | Police dispatch method, police dispatch equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116596274A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117095802A (en) * | 2023-10-17 | 2023-11-21 | 吉林大学 | Intelligent management system and method for accompanying personnel |
-
2023
- 2023-06-14 CN CN202310700886.XA patent/CN116596274A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117095802A (en) * | 2023-10-17 | 2023-11-21 | 吉林大学 | Intelligent management system and method for accompanying personnel |
| CN117095802B (en) * | 2023-10-17 | 2024-01-26 | 吉林大学 | Intelligent management system and method for accompanying personnel |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107222865B (en) | Communication swindle real-time detection method and system based on suspicious actions identification | |
| KR102288249B1 (en) | Information processing method, terminal, and computer storage medium | |
| CN112365894B (en) | AI-based composite voice interaction method and device and computer equipment | |
| CN108428446A (en) | Audio recognition method and device | |
| CN109543020B (en) | Query processing method and system | |
| CN109902957B (en) | Data processing method and device | |
| CN108447471A (en) | Audio recognition method and speech recognition equipment | |
| CN109872162A (en) | A kind of air control classifying identification method and system handling customer complaint information | |
| CN111428028A (en) | Information classification method based on deep learning and related equipment | |
| CN111666400B (en) | Message acquisition method, device, computer equipment and storage medium | |
| CN108010289A (en) | A kind of internet alarm method and system based on Application on Voiceprint Recognition | |
| CN116596274A (en) | Police dispatch method, police dispatch equipment and storage medium | |
| CN109492221A (en) | A kind of information replying method and wearable device based on semantic analysis | |
| CN114007131A (en) | Video monitoring method and device and related equipment | |
| CN106878347A (en) | Information processing method, system, mobile terminal and server | |
| CN115146124A (en) | Question-answering system response method and device, equipment, medium and product thereof | |
| CN115440221A (en) | Vehicle-mounted intelligent voice interaction method and system based on cloud computing | |
| CN117351336A (en) | Image auditing method and related equipment | |
| CN115238799A (en) | AI-based random forest malicious traffic detection method and system | |
| CN109710756B (en) | Document genre classification system and method based on semantic role labeling | |
| CN107133299B (en) | Fire-fighting answer method, mobile terminal and readable storage medium storing program for executing based on artificial intelligence | |
| CN113821587A (en) | Text relevance determination method, model training method, device and storage medium | |
| CN111210827B (en) | Method and device for responding to alarm, electronic equipment and readable storage medium | |
| CN115905187B (en) | Intelligent proposition system oriented to cloud computing engineering technician authentication | |
| CN116662522A (en) | Question answer recommendation method, storage medium and electronic equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB02 | Change of applicant information |
Address after: Room 233, 2nd Floor, Building 1, No. 2 Jinyuan Road, Daxing District Economic Development Zone, Beijing, 102627 Applicant after: Nancheng Yunqu (Beijing) Information Technology Co.,Ltd. Address before: Room 233, 2nd Floor, Building 1, No. 2 Jinyuan Road, Daxing District Economic Development Zone, Beijing, 102627 Applicant before: Nancheng Yunqu (Beijing) Network Technology Co.,Ltd. |
|
| CB02 | Change of applicant information |