181results about How to "Reduce complaint rate" patented technology

The invention discloses an intelligent collection method for overdue bill, the invention relates to a device, a computer device and a storage medium. The method comprises the following steps of: according to a customer risk grade dividing strategy, dividing the overdue customer information of an incoming case to obtain a list of overdue customers, wherein the list of overdue customers comprises overdue customer information and an overdue type identification corresponding to the overdue customer information; Obtaining a collection call template corresponding to an overdue type identifier, adding overdue customer information to a position corresponding to the collection call template, and obtaining a target collection call; The telephone platform is called to automatically dial the contact number of the overdue customer and convert the target collection call into the target collection voice, so as to obtain the overdue collection result of the overdue customer's feedback to the target collection voice. Improve collection efficiency, reduce collection costs and customer complaints due to improper use of collection strategies.

The invention discloses a deep learning-based electric power suspected complaint work order recognition method, and relates to complaint recognition method for electric power customer services. At present, electric power suspected complaint work orders are artificially recognized, so that the efficiency is low, the timeliness is bad and human resources are wasted. The method comprises the following steps of: 1) deep learning model configuration; 2) complaint feature label extraction; 3) complaint sample formatting; 4) model learning and training; 5) suspected complaint recognition; and 6) suspected complaint classification. According to the method, a series of work such as data cleaning and sorting, complaint tendency word extraction, data modeling and optimization, sample iterative learning and training, machine learning and prediction and the like can be carried out on complaint lists through a deep learning technology, so that deep learning-based intelligent recognition and classification of suspected complaint work orders are realized, the intelligent work experience is enhanced and the service quality control work efficiency is improved.

The invention discloses an electric charge sensitivity assessment method based on logistic regression. According to the method, electric charge sensitivity models are respectively established for high-voltage users, low-voltage non-resident customers, and resident customers by regarding customer sensitivity as the entry point. The main steps comprise: collecting modeling indexes from a plurality of dimensions including customer basic information, electricity consumption information, and payment information etc., screening variables by employing information values (IV) and related coefficients, grouping the variables based on the optimal grouping algorithm and the optimal clustering algorithm, performing conversion of weight of evidence (WOE), establishing the customer electric charge sensitivity assessment model by employing a logistic regression algorithm, constructing a standard scoring card which is easily understood and implemented according to model parameter estimation values, and finally determining the weights of the variables through an advantage analysis method. According to the method, data support is provided for development of accurate marketing and differentiated service by departments of electric power marketing and customer service through recognition of customers with high electric charge sensitivity, the overall satisfaction degree of the customers is improved, and the customer perception is improved.

The invention discloses an electricity fee sensitive user analysis method based on stacking and bagging algorithms. The electricity fee sensitive user analysis method comprises steps of using an optimal zone algorithm to solve problems of nonequilibrium and intolerance and construct a training set and a testing set according to electricity fee sensitive classification targets and on the basis of multiple core business indexes of an electricity fee sensitive original data table, further subdividing related core business structuralized characteristics based on electricity fee sensitivity, using a stacking method to construct an electricity sensitive primary model on the basis of nonstructural text characteristics of particles and word frequency statistics, generating extended stacking characteristics for each sample, combining characteristics of the two as an integral input, using a bagging algorithm and a vote algorithm to construct an electricity fee sensitivity secondary model on the training set and the testing set, using a trained model to perform prediction on a verification set and verifying during practical business. The Electricity fee sensitive user analysis method based on stacking and bagging algorithms can improve mastery degree of an electric power company about electricity fee sensitivity of the user, and is beneficial for providing differential and specific quality power supply service.

The invention relates to a power distribution network reconstruction demand model building implementation method based on customer demands. The method comprises the steps that firstly, data acquisition is performed; secondly, an automatic rating model is built, wherein index constructing, weight determining and score mapping are sequentially performed; thirdly, reconstruction demand rating is performed, wherein characteristic data of a target region are extracted, the characteristic data are automatically calculated and scores are obtained on the basis of the automatic rating model, interval division is performed on the scores, then the level of the corresponding transformer district reconstruction demand is obtained, the characteristic data comprise transformer districts, 95598 work orders and marketing archives; fourthly, rating outputting is performed, wherein the transformer district reconstruction demand rating result is stored in a corresponding database for query; fifthly, the rating result applied to work of a power supply customer service center and an electric power operation and maintenance department. According to the power distribution network reconstruction demand model building implementation method based on the customer demands, distribution network transformer districts in which service hidden hazards exist can be specially focused; accordingly, capital investment of distribution network reconstruction is guided, the service hidden hazards are eliminated in time, the customer complaint rate is decreased, and the customer satisfaction degree is improved.

The invention discloses a method for rating complaint willingness of a logistics client. The method comprises the following steps: acquiring client information and client complaint content, arranging an index, and performing ordering, wherein the client information comprises client ID, complaint order numbers, complaint time and evaluation information given by logistic work personnel for clients; according to a time sequence, dividing complaints of the clients into different time periods; calculating complaint probabilities of the clients in the different time periods, and further obtaining complaint probability values of the clients; according to the client information, dividing the client information into different dimensions, distributing weights according to importance of each dimension, and calculating emotion grade values of the clients by use of text, voice and face emotion identification technologies; according to the complaint probability values and the emotion grade values of the clients, calculating complaint willingness total scores of the clients; and according to the complaint willingness total scores of the clients, providing different dispensing priority values for packages, and thus the complaint rate can be substantially reduced.

The invention discloses an information management method. The method includes: dividing all customers into groups according to the first information and the second information in the customer information data pool; business recommendation table; when receiving the notification that the terminal held by the customer logs in to the external device, recommend business to the customer according to the business recommendation table corresponding to the group to which the customer belongs; collect the results of the business recommendation, and The collected results update the business referral form. The invention also discloses an information management platform and system at the same time. By adopting the technical solution of the present invention, after targeted recommendations are made to customers, the recommendation results can be automatically evaluated; through the evaluation results of recommendations from other similar customers of the same category, the recommendation algorithm can be continuously optimized, thereby improving the accuracy of business recommendations; Eliminate customers' aversion to business recommendations and reduce customer complaints.

The invention discloses a system and a method for monitoring the whole process of customer power purchase service. A bypass integrated payment access management platform system proxy server, a prepayment electricity selling system proxy server, a marketing business application system proxy server and an electricity consumption information acquisition system proxy server are built; each proxy server is responsible for detecting system interfaces, logs and database data; the data is analyzed according to the service logic and summarizing the data; an internal memory database technology is used for storing the analysis summary data in a client power purchase whole-process monitoring large number server (D3BASE) in real time, whether a system gives an alarm or not and the alarm influence service range are judged according to the content of an alarm threshold value, an abnormal label library and the like preset in advance, and a WEB application page and a short message interface are provided for a client to be displayed.

The invention provides an operation method for controlling the reliability of an air conditioner, a control system and the air conditioner. The operation method for controlling the reliability of theair conditioner comprises the following steps that whether a temperature sensor fails or not is judged according to a temperature signal sent by the temperature sensor, if the temperature sensor fails, the air conditioner continues to acquire real-time outdoor environment temperature, the air conditioner automatically matches each operation parameter of the system corresponding to the outdoor temperature according to the real-time outdoor temperature, and continues to operate according to each operation parameter of the system after automatic matching; and otherwise, normal mode operation is continued. The operation method for controlling the reliability of the air conditioner can continuously control normal operation of the air conditioner after the temperature sensor fails, so that the normal use of people is met.

The invention discloses an air conditioner with a refrigerant leakage display function. The air conditioner comprises an electric control device and a display device, wherein the electric control device is provided with a detection module, the detection module is provided with a detection component for detecting the operation state of the air conditioner, and the display device is provided with a refrigerant display component for displaying the quantity of remaining refrigerants or leaked refrigerants of the air conditioner. The invention further discloses a display control method of the air conditioner. By means of the technical scheme, the quantity of the remaining refrigerants or the leaked refrigerants of the air conditioner can be displayed in real time, a user can know whether the refrigerants of the air conditioner are leaked according to the display condition, and accordingly faults of the air conditioner caused by leaking of the refrigerants can be prevented in time; more serious accidents are avoided later, and the complaint rate of the user can be effectively reduced.

The invention relates to an autonomous behavior-based product marketing support system and method, is a support system which can provide marketing management for a plurality of products at the same time and initiatively provides a novel product marketing mode which is suitable for ordinary users to participate extensively and has great influence. Except for relevant network elements and various product marketing platforms of an original telephone switching network, the product marketing support system is also provided with the autonomous behavior-based product marketing support platform. The system has the following main functions: mainly adopting user autonomous promotion in a short message or network mode to provide parallel support for multi-product and multi-region marketing; and meanwhile, ensuring that transmission or marketing tasks of each product or service are independent to realize noninterference among promotion and transmission tasks. Moreover, the marketing support system and method can strictly limit participating populations and regions and provide real-time marketing awards for participating users. Meanwhile, marketing personnel can learn personal marketing performance in real time, thereby furthest arousing the working enthusiasm of marketing personnel.

Belonging to the field of electric power technology, the invention specifically relates to a wireless transmission anti-missort tester and a working method thereof. The wireless transmission anti-missort tester includes a transmission device and a receiver, the transmission device is equipped with a power supply module, an ultra narrow band frequency selection module, a transmission channel and aground wire terminal. The ultra narrow band frequency selection module and the transmission channel are in parallel connection, the ground wire terminal is connected to the ultra narrow band frequencyselection module, and the output end of the ultra narrow band frequency selection module is connected to the input end of an amplification circuit. The tester provided by the invention can carry outdetection positioning without enter a household, and omits tedious access procedure. The tester has the advantages that: the time for correction detection of a user meter is short, the work efficiencyis improved, unified inspection is carried out on a plurality of electric meters in one meter box, and the operation is simple and convenient.

The invention relates to the technical field of cleaning machines, in particular to a tunnel type ultrasonic bottle cleaning machine. The tunnel type ultrasonic bottle cleaning machine comprises a machine case filled with a cleaning fluid, wherein a bottle inlet and a bottle outlet are formed in the machine case; a transmission device capable of circulating rotation is arranged in the machine case; a plurality of bottle carrying boxes are arranged on the transmission device; a bottle feeding mechanism for pushing bottles to be cleaned into the bottle carrying boxes is arranged at the bottle inlet; a bottle discharging mechanism for pushing the cleaned bottles out of the bottle carrying boxes is arranged at the bottle outlet. The bottle feeding device adopts a crank and rocker mechanism, movement dead points of a four-bar mechanism are overcome, the bottles are stably and reliable fed, a bottle discharging device adopts a connecting rod to receive the bottles, and the bottles are firstly discharged by a cushion pad and then fed by a bottle receiving claw, so that the tunnel type ultrasonic bottle cleaning machine works stably. A plurality of cleaning working procedures are set in the machine case, the cleaning requirement of the bottles is met, and the work efficiency is high.

The invention provides a PON optical communication terminal state general survey instrument. The PON optical communication terminal state general survey instrument is arranged at the rear end of an optical branching device of an optical distribution network (ODN) and is close to subscriber terminal equipment, and the PON optical communication terminal state general survey instrument comprises an optical receiving unit, an optical path distributing unit, a computing unit, a determination unit and a control unit. The PON optical communication terminal state general survey instrument provided bythe invention has the advantages that firstly the PON terminal state detection instrument according to the invention can carry out integrated network quality detection on all the circuits under a portof the branching device by being close to the rear end of the optical branching device of a subscriber without entering indoor space of the subscriber, so that high efficiency and timeliness of network maintenance are improved; and secondly, the PON terminal state detection instrument according to the invention adopts pass-type detection, and the subscriber is not influenced even if in Internet use, so that satisfaction degree of the subscriber in use is improved, and complaint rate of the subscriber is reduced.

The invention provides a multifunctional unmanned distribution flying cart and a working method thereof; the multifunctional unmanned distribution flying cart comprises a fuselage shell, wherein the fuselage shell is provided with a support shaft which is axially and fixedly connected; the outer end of the support shaft is hinged with a foldable side shaft; a propeller driven by a propeller motorto rotate is fixed at the end part of the foldable side shaft; a storage box body used for storing articles is arranged below the fuselage shell; a pair of bottom doors capable of opening and closingare arranged at the bottom of the storage box body; wheels are arranged below the storage box body; according to the multifunctional unmanned distribution flying cart, the change of a flying state anda ground walking state can be achieved through the change of the direction of the foldable side shaft; meanwhile, the transportation is achieved by utilizing the storage box body of the articles; thefunctions of unmanned unloading and self-taking of the articles can be achieved through cooperation; the multifunctional unmanned distribution flying cart is driven by a belt wheel and a gear; according to the multifunctional unmanned distribution flying cart, the service quality of express delivery industry can be improved, the delay rate of express items is reduced, and the express delivery issafer, more reliable and more rapid.

The invention discloses a charging method, which is applied to networks where identification of status and identification of position are separated. The charging method includes the following steps: when an access node is registered at an authentication node which a user belongs to, user access charging parameters are sent to the authentication node, wherein the user access charging parameters comprise identification of status of the user, identification of position and mode of access of the access node accessed by a user terminal, identification of an operator of the access node and identification of geographic position of the user; the authentication node obtains charge rate for the identification of status of the user in the access mode from the charging node through the access charging parameters and returns the charge rate to the access node; and the access node charges the user in the access mode according to the charge rate. The invention further discloses an access device. Due to the access device, no matter where a user terminal roams, an access stack node (ASN) can charge correctly according to charge rate.

The invention discloses a risk management model training method. The method comprises: a first historical data group of known results is obtained, wherein the first historical data group includes account information of an abnormal user; according to the type of the account information, a data mining model with a highest fitting degree is selected as a first stage model; dimensionality of account information in first historical data is adjusted by using the first stage model to generate a second historical data group; and after the first stage model is trained by using a first data set in the second historical data group, the first stage model is verified by using a second data set in the second historical data group and a risk management model is selected based on a verification result, wherein the risk management model is used for identifying an abnormal user in telephone bill payment. In addition, the invention also discloses a risk management model training apparatus. With the method and apparatus, a complaint rate can be reduced.

The invention discloses a method for evaluating the transmission-generated abnormal sound of a vehicle hydraulic shock absorber, which comprises the following steps of: S1, using an iteratively-loaded sinusoidal signal as a displacement excitation of a test bench moving head to simulate the driving of a car on a rough road surface; S2, setting a sound pressure sensor; S3, using the score rating method to make subjective evaluations of the near-field noise radiated by the absorber itself, and based on this, evaluating the transmission-generated abnormal sound level of the absorber; S4, dividing the collected near-field radiation noise frequency of the shock absorber into blocks, calculating the linear sound pressure level of the noise signal in each sub-frequency band, finding the characteristic sub-frequency band with the highest correlation with the subjective evaluations, and based on this, determining the characteristic frequency band range for the shock absorber's transmission-generated abnormal sound; and S5, making objective evaluations on the shock absorber's transmission-generated abnormal sound indicators. The method of the invention directly utilizes the radiation noise of the shock absorber to carry out the evaluation of the transmission-generated abnormal sound of a shock absorber, which can reflect the level of the transmission-generated abnormal sound of the shock absorber simply and truly.

The invention discloses a quick connector device for a flexible medium-voltage cable. The device comprises a middle socket and two cable terminal plugs, wherein one ends of the two cable terminal plugs are respectively connected to the ends of the two cables, and the other ends of the two cable terminal plugs are detachably connected to the two ends of the middle socket respectively. According tothe quick connector device, through the two cable terminal sockets and the middle socket, a standardized cable connector is achieved, and a connection between a ring main unit and the cable becomes very convenient and efficient; compared with a traditional copper pigtail connection mode, the connection speed of the connector device is increased by tens of times, a problem that the connection efficiency is low in the traditional mode is effectively solved, the on-site rapid connection is realized, the all-weather working environment is adapted, the efficiency of installation and power transmission can be effectively improved, a user complaint rate can be reduced, the connector is normalized, and the device adapts to connections of the cables of different manufacturers and is easy to installand convenient to match.

The embodiment of the invention discloses a data deduplication method and device, computer equipment and a storage medium. The data deduplication method comprises the steps: obtaining a data access request, and extracting a feature field in the data access request; cleaning the feature fields, and normalizing the cleaned feature fields; splicing the feature fields to generate a feature field combination, and compressing the feature field combination by using a hash algorithm; identifying the compressed feature field based on a preset database cluster, and judging whether the feature field is arepeated field or not according to an identification result; and if the feature field is a repeated field, storing the feature field into a preset exception handling queue, otherwise, outputting a prompt message which is used for prompting that the feature field is a normal field. The invention provides a data deduplication method based on data processing, capable of reducing the consumption of alarge amount of repeated data for database resources, saving the memory space of a database, and improving the reputation of an enterprise.

The invention discloses a power failure sensitivity early warning method and device based on heterologous cross regression analysis. A sample data set is formed, related performance measures of the variable factors are evaluated by selecting the variable factors, a certain threshold value is set for the variable factors according to the related performance measures of the customer power failure satisfaction degree of the variable factors, and early warning is conducted on customer power failure complaints. The power failure sensitivity early warning method and device based on heterologous cross regression analysis are helpful for power enterprises to accurately identify customers with high power failure sensitivity, reduce the probability of complaints caused by electricity charge errors of the customers, and achieve the purpose of integrally improving the satisfaction degree of the customers.

The present invention discloses a method for limiting short message sending number of times. The method comprises the steps of collecting a current short message input by a user; matching the current short message with a historical short message in a cache system; and if the matching succeeds, determining whether matching number of times reach limited number of times, wherein if the matching number of times reach the limited number of times, the current short message is not sent. Through adoption of the method, the same or similar content is prevented from being sent repeatedly, and the same phone number is prevented from sending short messages of the same or similar content for multiple times so as not to cause bad impact on a mobile phone user, so that the short message validity is improved, the user experience is improved, and the complaint rate is lowered.