25 results about "Call duration" patented technology

The embodiment of the invention provides a call quality detection method and device, electronic equipment and a storage medium, and is applied to the technical field of communication. The method provided by the embodiment of the invention is applied. The method comprises the following steps: acquiring a single call MOS value, a single call packet loss rate, single call average jitter, single call average delay, single call maximum jitter, a single call packet loss rate peak value, a single call codec type, a single call network type and a single call duration of each call of a user in a preset historical duration; and calculating the first data and constructing a user feature vector to obtain the user feature vector representing the call quality of the user. And respectively analyzing the feature vectors of the users through a K-means clustering algorithm and a DBSCAN clustering algorithm, identifying abnormal call users, and clustering the users, so that the users with the same or similar call quality are clustered in the same cluster. Therefore, the long-term call quality of the user is judged.

The invention discloses a multi-source data business travel service collaborative decision-making method and system, and relates to the technical field of business travel services, and the technical scheme is characterized in that the method comprises the steps: collecting business travel demand scene information of a current user, and calculating the resource matching complexity according to the business travel demand scene information; comparing and judging the resource matching complexity to obtain a first-level collaborative decision signal or a continuous analysis signal; if the resource matching complexity is compared and judged as a continuous analysis signal, extracting historical behavior data of the user, and constructing a resource relation model associated with the current scene according to the historical behavior data; analyzing calling time sequence characteristics of the user to the resource nodes based on the resource relation model; wherein the calling time sequence characteristic comprises an abnormal calling duration proportion and a continuous abnormal calling frequency; generating a time sequence exception index according to the exception call duration ratio and the continuous exception call frequency; the effect is that resources and service power can be flexibly allocated.

The invention provides a double-handle dispatching desk dialing display and control method and device and a storage medium, and the method comprises the steps: when a dispatching desk is started, loading a system configuration file and a handle drive program, detecting the connection state of at least two independent handles through a hardware interface, displaying a handle selection list on a main interface, and generating an adaptive layout for each handle icon; if dialing succeeds, information such as call duration and the number of the opposite side is displayed on a display screen of the first handle, functional keys such as on-hook and keeping are activated on the second handle, and if dialing fails, the display screen displays specific failure reasons and provides shortcut buttons for one-key redial and input emptying; through an interface layout optimization algorithm, high-frequency operation including dialing and mode switching is mapped to handle entity keys, interface level jump is reduced, a system automatically memorizes operation habits of a user, historical configuration is preloaded during next starting, and an operation path is shortened.

The invention provides a user risk assessment method and device, equipment, a medium and a product, and belongs to the technical field of communication security, and the method comprises the steps: obtaining historical risk data, historical communication behavior data, historical consumption behavior data and historical associated information data of a target user in response to a network access request of the target user; if the historical risk data represents that the target user is a non-risk user, performing feature extraction on the historical communication behavior data, the historical consumption behavior data and the historical associated information data; and inputting the obtained age, occupation, roaming place, package monthly fee, monthly average call duration, monthly average call times and monthly traffic usage amount of the target user into a pre-trained risk assessment model to obtain a risk coefficient output by the risk assessment model, and obtaining an assessment result of the target user according to the risk coefficient. According to the invention, risk assessment is carried out when the target user initiates the network access request, pre-assessment is realized, and risk users are prevented from implementing risk behaviors based on communication means.

The invention discloses a data processing method and device, equipment, a storage medium and a program product, and relates to the technical field of artificial intelligence. Obtaining service session data; determining N service links and starting and ending time of each service link through a large language model according to the service session data; according to the starting and ending time of each service link, determining a time-consuming service link in the N service links, the service duration of the time-consuming service link being greater than or equal to a preset service duration; according to stage session data corresponding to the time-consuming service link in the service session data, time-consuming root cause data is determined, and the time-consuming root cause data comprises at least one of the following items: a specialist factor and a non-specialist factor. Therefore, the problems that traditional manual analysis depends on experience judgment and the root cause is fuzzy are effectively solved, and accurate positioning and root cause tracing of the customer service call duration problem can be realized through intelligent analysis of the service session data by the large language model, so that the requirements of improving the service efficiency and the service quality are met.

This invention discloses an intelligent emergency call method and system, applicable to any user group in an emergency situation, including the elderly, women living alone, and patients with chronic diseases. It can run on various smart terminals such as smartphones, smartwatches, and smart bracelets. After a user triggers the call with one click, the system uses a multi-round cyclic dialing strategy to dial multiple emergency contacts sequentially, with subsequent rounds only dialing those who did not answer. Accurate answer determination is achieved by obtaining the actual call duration (zero indicates no answer). Simultaneously, it automatically integrates the GPS reverse geocoding address with the user's health record into a structured emergency call notification message and sends it. The entire process includes voice guidance, and all interactive nodes employ a countdown default continuation mechanism to ensure the process is not interrupted due to user inability to operate. This invention solves the three major defects of existing technologies: slow response, frequent misjudgments, and incomplete information. The entire process is triggered with one click and executed automatically.

The present application relates to the technical field of data analysis, in particular to a kind of intelligent call data analysis and prediction method, comprising the following steps, through communication equipment extraction call information generation node sequence and weight matrix, sliding detection statistical compression degree score extraction sub-sequence, joint feature model assesses path type, scanning frequency sequence extracts event-driven time stamp generation behavior label, based on hidden Markov model prediction behavior evolution result.In the present application, by time series adjacency clustering, call events are integrated into path node sequence, normalized call duration is generated weight distribution, sliding three-node algorithm identifies high-frequency compressible sub-sequence, fuses node density path length change and weight concentration degree to construct joint discriminant model, classifies call path type, dynamic frequency variable window extracts frequency difference to enhance behavior recognition sensitivity, and hidden Markov model dynamically predicts call behavior evolution trend based on state transition probability.

The application relates to the technical field of information tracing, in particular to a data security sharing method and equipment of a cross-network exchange platform, which comprises the following steps: determining a first abnormal characteristic value based on the distribution of deviated characteristic values of external requests of all Span data under any service node in each time period; determining a second abnormal characteristic value based on the change trend of the calling duration of all Span data under any service node in each time period, and combining the first abnormal characteristic value to determine a sampling adjustment coefficient of any service node in each time period, so as to adjust the sampling rate of Span data under any service node in the next time period of each time period. The application solves the problem that abnormal data is easy to be missed in the traditional method by dynamically adjusting the link tracking sampling rate, and improves the efficiency and accuracy of fault positioning of the cross-network exchange platform.

The invention relates to the technical field of resource allocation, in particular to an office resource allocation method based on the Internet of Things, which comprises the following steps: acquiring a task flow and a resource calling record, identifying a task dependence flow, extracting a resource type, a use time period and calling frequency, constructing a resource demand structure list, calculating resource occupation, and marking a high occupation flow. Generating a conflict area list, screening high-conflict requests, executing resource rearrangement, correcting calling duration, and generating allocation information. According to the method, the resource demand distribution structure is constructed by extracting the resource type, the use time period and the calling frequency in the task process, the resource allocation conflict area is identified, the combined screening is executed, the resource calling urgency degree is calculated in combination with the task priority and the resource utilization rate, and the high-conflict request starting point is effectively positioned; by eliminating the calling records with long resource idle period and low task level, the reasonability of task scheduling and the stability of office resource supply are enhanced.

Disclosed in the present application is a video call processing method. The video call processing method is applied to a cloud album server, and comprises: acquiring scene characteristic information of a first client that is sent by the first client; searching a cloud album for a target scene image, which matches the scene characteristic information and a possible call duration of a video call; generating an image video stream on the basis of the target scene image; and sending the image video stream to a video stream receiving end, such that the video stream receiving end replaces the video stream, which is sent to a second client, with the image video stream, and the video stream receiving end comprises at least one of an instant messaging server and the first client.

The invention relates to the technical field of user portraying, and discloses a user portraying method and system based on call interaction behaviors. The system comprises a call data acquisition module, a behavior feature extraction module, a user portrait construction module, a collection strategy generation module and a strategy execution feedback module. The call data acquisition module acquires metadata such as call recording data and call duration in real time; a behavior feature extraction module extracts interactive behavior features of dimensions such as tone intensity, response speed and language content tendency; the user portrait construction module continuously updates a user portrait label set according to the dynamic characteristics; the collection strategy generation module formulates adaptive personalized collection strategies for users with different characteristics in combination with the updated label set; when the strategy execution feedback module executes the strategy, effect data such as user repayment response and communication adaptability are synchronously collected. According to the system, through full-process closed-loop operation, the collection business process is optimized, and the user portrait accuracy and the collection strategy pertinence are improved.

The embodiment of the invention provides a call center capacity evaluation method, a call center capacity evaluation device and related equipment. In the embodiment of the invention, the maximum audio concurrency support capability of a single server and the resource consumption ratio of video to audio can be obtained, and the video traffic proportion of a call center and the call duration ratio of the video to the audio can be obtained; determining the maximum concurrency support capability of a single server under the mixed telephone traffic through a first calculation model, and predicting the total concurrency demand of the call center under the mixed telephone traffic through a second calculation model according to the number of calls per second, the audio call duration, the video telephone traffic proportion and the call duration proportion of the call center; and based on the total concurrency demand and the maximum concurrency support capability of the single server under the mixed telephone traffic, determining a preliminary server demand quantity of the call center, and based on the performance difference correction factor, correcting the preliminary server demand quantity to obtain the corrected server demand quantity of the call center.

The application belongs to the technical field of access control management, and discloses an entrance access control management method, device and equipment. The method comprises the following steps: collecting and identifying the identity information of a person passing through; determining target inspection items of the person passing through; determining the optimal inspection order based on the calling duration corresponding to each target inspection item; determining the target inspection item with the longest calling duration based on the optimal inspection order; preferentially initiating an inspection request of the target inspection item with the longest calling duration; sequentially calling other target inspection items in the optimal inspection order while waiting for the inspection result of the target inspection item with the longest calling duration; and determining the permission inspection of the person passing through after receiving the inspection results of all the target inspection items. The device comprises an identity identification module, an item determination module, an ordering module, a calling module and a determination module. The application optimizes the determination strategy of the inspection items and dynamically adjusts the inspection strategy, thereby improving the access control efficiency.

The invention relates to the technical field of customer relationship management, in particular to an edge-deployed customer loss prediction method and system based on machine learning, and the method comprises the following steps: obtaining historical behavior data, attribute data and service use record data of a target user, the data comprises but is not limited to communication frequency, call duration, internet traffic, bill amount, complaint record and package change information; the method has the beneficial effects that an integrated model (such as XGBoost and LightGBM) based on a gradient boosting decision tree is adopted, and nonlinear features and complex interaction relationships in customer churn behaviors can be effectively modeled. Compared with a traditional logistic regression model, the method has the advantage that the prediction precision on an actual operation data set is remarkably improved. Through K-fold cross validation and a regularization mechanism (such as L1 / L2 constraint), the volatility of the model in different time periods or different service scenes is remarkably reduced, and the robustness of data disturbance is improved.

The invention provides a predictive outbound call-to-artificial resource scheduling method and system, and belongs to the technical field of task scheduling. According to the method, through ringing duration box separation processing, the answering probability difference (for example, the answering rate of ringing 15S is remarkably higher than 2S) of different ringing duration stages can be quantified, and therefore the accuracy of calculating the predicted answering number at the next moment can be improved. In consideration of a non-linear relationship between emotional fluctuation and the hanging rate, for example, the hanging rate of a negative emotional call is higher than that of a positive call. Therefore, the system can dynamically capture the complex behavior pattern through cross analysis of the call duration interval and the emotion label, so that the prediction precision of predicting the on-hook number at the next moment can be effectively improved. And the number of idle seats at the next moment is accurately calculated through the dynamically predicted answering number and hang-up number, and the call-out task amount is dynamically adjusted, so that call discarding and queue congestion can be effectively avoided, and the user experience and the call-out efficiency are greatly optimized.

This invention relates to the field of dynamic load allocation technology, and discloses a method and apparatus for dynamic load allocation in power systems. The method includes acquiring original sampled signals and unit response parameters including inertial time constants, regulation rate limits, and standby call durations; performing interpolation and alignment to obtain a preliminary load time series; cleaning outliers based on the preliminary load time series to obtain a cleaned load sequence; iteratively fitting the cleaned load sequence based on matching similarity judgment to obtain short-term trend characteristics; performing anomaly pattern matching based on the short-term trend characteristics to obtain fluctuation confirmation indicators; performing trajectory correction based on the fluctuation confirmation indicators and unit response parameters to obtain a corrected prediction sequence; iteratively optimizing commands based on feasibility judgment based on the corrected prediction sequence to obtain feasible commands; and iteratively adjusting commands based on feasibility commands to obtain a final command sequence. This method can improve the safety and efficiency of power grid operation.

1.The name of the design product: the call module body of the smart conversation graphical user interface of an electronic device. 2.The use of the design product: an electronic device. 3.The design points of the design product: the part of the graphical user interface in the electronic device. 4.The picture or photo that best indicates the design points: change state diagram 2. 5.The use of the graphical user interface: the graphical user interface is used for conversation. The part of the graphical user interface that is protected by the request is used for voice or video call. By clicking the "video call" icon above the input box at the bottom of the graphical user interface, the graphical user interface changes from the main view to change state diagram 1, entering the smart voice interaction interface; the user clicks the red end call button in the control bar at the bottom of the graphical user interface, the graphical user interface changes from change state diagram 1 to change state diagram 2, exits the smart voice interaction interface and returns to the conversation page, a "call has ended" label appears at the top of the interface, and the call duration is displayed in the label, at the same time, the call content (such as text, file, etc.) is displayed in the conversation flow, the "open video call" mark is displayed at the top of the conversation flow, and the "call has ended" mark is displayed at the bottom of the conversation flow. 6.Other circumstances that need to be explained: the dashed line in each view shows part of the graphical user interface, which does not constitute the part protected by the design.

A method for identifying accounts used to make noncompliant calls (e.g., robocalls or voice spam calls) includes obtaining, by a communication services platform, one or more metrics for voice calls associated with a first identifier of a first communication endpoint. The one or more metrics include an average call duration for the first identifier or a predetermined signaling protocol code. The method includes determining, based on the metrics, a risk score associated with the first identifier. The method includes performing, based on the risk score, a corrective action with respect to an account of the communication services platform. The account can be associated with the first communication endpoint.

The invention relates to the technical field of operation management, and discloses an SOP task and daily report fusion system and method based on AI intelligent analysis. According to the invention, standardized SOP tasks and daily reports are deeply fused to form a structured daily report worksheet, and a multi-dimensional and closed-loop performance evaluation mechanism is established based on AI intelligent automatic comparison and analysis: firstly, based on sales data comparison of an experiment and a historical period, actual pulling of process optimization to business return is visually verified; secondly, by combining total working hours and unit working hour output analysis, pseudo growth caused by overtime investment is stripped, and it is ensured that performance improvement comes from real improvement of execution efficiency; and thirdly, double communication input indexes of the invitation call duration and the online typewriting quantity are adopted, so that performance illusion caused by contact of more customers is eliminated, and the increment performance is ensured to really benefit from an efficient and standardized execution path.

A method for call processing includes: receiving a first incoming telephone call from a first person; retrieving historical information that relates to the first person; forecasting, based on the historical information, a next customer state that relates to the first person; and estimating, based on the forecasted next customer state, a potential duration of the first incoming call. The estimated call duration is then usable for assigning a priority to the first incoming call and determining an order of handling the first incoming call with respect to each of a plurality of other incoming calls based on the assigned priority.

The application provides a tariff calculation method and device, electronic equipment and storage medium, and relates to the technical field of data processing. The method comprises the following steps: collecting call signaling data of user call behavior and generating real-time usage data; updating the cumulative call duration sequence of the user based on the real-time usage data and storing the cumulative call duration sequence into a time series database; when receiving a tariff query request, determining the corresponding user identifier and query time span, obtaining the tariff package data effective within the query time span in the relational database based on the user identifier, and dividing the query time span into at least one billing interval based on the tariff package data; for each billing interval, calling the cumulative call duration corresponding to the start and end time nodes of the billing interval from the time series database and calculating the difference to obtain the call usage within the billing interval; and calculating the total tariff within the query time span based on the call usage corresponding to each billing interval and the call unit price of the tariff package into which each billing interval is divided. The application improves the tariff query efficiency.

This application provides a video call method and an electronic device. The method includes: a first device supporting one-way video call function sending a request to a second device to establish a one-way video call; after establishing a call connection with the second device, the first device timing the call duration and receiving image data from the second device; in response to not receiving image data after the call duration reaches a first predetermined time, the first device sending a request to the second device to establish a two-way video call; after sending the request to the second device to establish a two-way video call, the first device continuing to receive image data sent by the second device; in response to receiving image data sent by the second device, playing the image data sent by the second device, and updating the stored one-way call support information to include the network identifier of the second device and a function skip tag corresponding to the network identifier, wherein the function skip tag is used to indicate that one-way video call function is not supported.

The application discloses a harassment and fraud phone identification method and device, equipment and storage medium, relates to the technical field of data processing, and is used for improving the identification efficiency and accuracy of harassment and fraud phone. Including: based on the basic information of the target phone identifier, a plurality of basic parameters corresponding to the target phone identifier are determined, the plurality of basic parameters include at least one of the following: caller proportion, caller concentration, called concentration, average call duration and daily average traffic, the caller proportion is used to indicate the ratio of the calling times of the target phone identifier to the total calling times, and the caller concentration is used to indicate the proportion of the first type of number in the calling number of the target phone identifier; according to the plurality of basic parameters and the weight value corresponding to each basic parameter, the target parameter corresponding to the target phone identifier is determined; when it is determined that the target parameter is greater than the target threshold value, the target phone identifier is determined as the identifier of the harassment and fraud phone.

The invention relates to the technical field of voice processing, can be applied to business scenes of financial science and technology, medical health and the like, and discloses a voice communication method, device, equipment and medium based on customer group timbre preference. Tone preference analysis is carried out for different guest groups and an analysis result is generated, a matching relationship between guest group labels and sound models is established based on the analysis result, the matching relationship is inquired according to the guest group labels and a target sound model is acquired when a voice call is initiated, and voice content is generated by using the target sound model to complete the call. And collecting call effect data to optimize the matching relationship. Through matching of the customer group label and the sound model and dynamic optimization of the call effect data, self-adaptive combination of sound and customer preference is realized, the personalization and affinity of voice content are improved, the call duration is prolonged, and the response rate and the call-out capacity are improved.

The application relates to a service pushing method and device, equipment and a storage medium, and relates to the technical field of communication. The method comprises the following steps: obtaining a target call duration of a target user in a preset time period, service interaction data, traffic usage data, and promotion channel data of a target service in the preset time period; a target age range corresponds to the target service; the service interaction data is the number of times that the target user contacts the service; the promotion channel data is the number of times that the target user contacts the promotion channel of the target service; determining a target matching coefficient between the target service and the target user according to the target call duration, the service interaction data, the promotion channel data and the traffic usage data; the target matching coefficient is positively correlated with the target call duration, the service interaction data, the promotion channel data and the traffic usage data. Further, if the target matching coefficient is greater than or equal to a service matching threshold corresponding to the target service, the target service is pushed to the target user.