The invention provides an intelligent robot customer service marketing method and system based on a user tag system. The intelligent robot customer service marketing method comprises the steps of establishing a user tag database, wherein data information for a plurality of users is stored in the user tag database, the data information is stored in the user tag database in a form of category tags,and the data information at least includes contact ways of the users; sending a voice request having a corresponding marketing content to each of the users by using an intelligent robot having a voiceinteraction logical function according to a robot marketing rule that is set in advance and is based on the category tags; and automatically recording and analyzing a voice call record between the intelligent robot and the each of the users, and performing client will classification processing on call results. According to the intelligent robot customer service marketing method and system, the mechanization of telephone marketing can be implemented, the utilization rate of the time is improved, the manpower cost is saved and the sales performance is improved.

Intelligent robot customer service marketing method and system based on user tag system

The invention discloses an adaptive distributed system structure. By adaptively adjusting the scale of a specified microservice instance, the challenge of controlling the cost and keeping the qualityof the cloud service caused by a user demand change can be handled. In the adaptive distributed system structure disclosed by the invention, the inherent relationship between concurrency and performance is described from the perspective of a distributed cluster level, the adaptive distributed system structure has good model generalization ability, can be widely applied to distributed clusters withdifferent structures and behavior characteristics, and effectively reduces the work intensity of IDC operation and maintenance research personnel. In addition, the backbone algorithm of the adaptivedistributed system structure disclosed by the invention is derived from the improvement of the google deepmind team based on the Q Learning, the algorithm does not need to have correct input/output pairs, nor does it need to accurately correct the suboptimal behaviors, and the algorithm is more focused on online planning and needs to find a balance between exploration (in the unknown fields) andcompliance (existing knowledge).

Adaptive distributed system structure

The invention presents a service hotline broadcasting method and device, a piece of computer equipment and a storage medium. The method includes the following steps: acquiring first user information according to the entry number of a user entering a service hotline by telephone; inputting the first user information to a first-level prediction model for calculation to obtain a first probability value of each service in a service category that the user intends to handle, wherein the first-level prediction model is obtained by training a specified sample set based on a gradient lifting decision tree and a logistic regression algorithm; and broadcasting the services corresponding to the first probability values in turn according to the descending order of the first probability values. Throughthe method, the service demand of users entering a service hotline by telephone can be predicted, users can hear the services needed, the time for users to find a service menu can be shortened, and the user experience can be improved.

Service hotline broadcasting method and device, computer equipment and storage medium

The invention discloses an insurance recommendation method. The insurance recommendation method comprises the following steps of: after receiving an insurance recommendation request with client identification information and sent from a front end, finding feedback data of a client corresponding to the client identification information for marketing mode from a pre-determined first data source; iffinding the feedback data of the client corresponding to the client identification information for the marketing mode, analyzing the feedback data of the client for the marketing mode according to a preset marketing-mode preference judgment rule so as to determine the preferred marketing mode of the client; sending an insurance recommendation instruction for the marketing mode of the client to thefront end according to the preferred marketing mode of the client. The insurance recommendation method disclosed by the invention has the advantages that insurance recommendation can be carried out for the client according to the preferred marketing mode of the client, the experience effect of the client can be improved, potential intentional clients can be found and the business efficiency of business personnel can be improved.

Electronic device, insurance recommendation method and computer-readable storage medium

The invention discloses a population prediction method, comprising the following steps: (1) acquiring data related to population activity quantity in a predetermined time period in a predetermined spatial unit, wherein the data includes mobile phone signaling data, meteorological data and other obtainable population activity related data; (2) cleaning the mobile phone signaling data obtained in the step (1) to obtain fine data; (3) carrying out spatial matching and fusion on that multi-source heterogeneous big data, and carrying out spatial unit matching on the data of other influencing factors affecting the spatial distribution of population activity and the clean mobile phone signaling big data in the step 2) to unify the data aperture; (4) using the data import model to forecast the population; (5) outputting the results of population size and spatial distribution.

A population prediction method based on multi-source big data and a long-short-term memory neural network model

The invention provides an insurance service recommendation method and system based on classified customer behavior analysis. The method comprises the steps of obtaining identity information of a client, and determining the type of the client according to the identity information of the client; recommending corresponding service to the client according to the type of the client. The system comprises a client information obtaining module, a client type determining module and a service recommending module. According to the insurance service recommendation method and system based on the classified customer behavior analysis, the type of the client is determined according to the identity information of the client, the corresponding service is recommended to the client according to the type of the client, thus recommendation is conducted according to the user or the service similar to the service once used by the user, the service pushing is more targeted, and thus the serving efficient and quality are improved.

Insurance service recommendation method and system based on classified customer behavior analysis

The invention provides a scenic spot tourist amount prediction system comprising a data collection module, a data prediction module and a flow display module, wherein the data collection module is used for collecting predicted data; the data prediction module is provided with a passenger flow prediction model based on LSTM and is used for predicting a scenic spot tourist amount according to the predicted data collected by the data collection module; and the flow display module is used for displaying the scenic spot tourist amount data. The scheme can predict the tourist amounts within different time periods.

Scenic spot tourist amount prediction system

The invention provides a service recommending method and system based on data mining. The method comprises the steps of obtaining customer call information, wherein the customer call information comprises a phone number of a customer; determining a type of the customer according to the phone number of the customer; extracting corresponding customer data according to the type of the customer; and carrying out data mining according to the customer data and generating a service recommending menu. The system comprises a customer call information obtaining module, a customer type determining module, a customer data extraction module and a data mining module. According to the method and the system, the customer call information is obtained, systematic mining analysis processing is carried out on the customer call information, and finally the interested service recommending menu of the customer is generated, so customer service processes are reduced, the system working efficiency is improved, an operation process of the interested is simplified, the time of the interested is saved, artificial customer service workloads are reduced, and the customer service experience of the customer is improved.

Service recommending method and system based on data mining

The invention provides a distributed deep learning operation and maintenance system. The system comprises a resource management module, a resource monitoring module, a log collecting and analyzing module and a mode management module. The resource management module is used for managing and scheduling resources required for deep learning; the resource monitoring module is used for monitoring resource consumption of clusters and tasks; the log collecting and analyzing module is used for collecting system operation logs; the model management module is connected with the resource monitoring module and the log collecting and analyzing module and used for conducting model management according to the resource consumption and the system operation logs. According to the scheme, the operation and maintenance management efficiency can be improved.

Distributed deep learning operation and maintenance system

The invention provides a tourist hotspot event detection method and system. The method comprises the steps of presetting a tourist professional knowledge base and obtaining network text data; clustering hotspot vocabularies according to the network text data; matching the hotspot vocabularies and the tourist professional knowledge base, and calculating the similarity; and setting weights of hotspot vocabularies according to the similarity, and performing re-clustering to obtain tourist hotspot events. According to the scheme, a final hotspot clustering result is more accurate.

Chen Long

Papers

Insurance service recommendation method and system based on classified customer behavior analysis

Scenic spot tourist amount prediction system

Service recommending method and system based on data mining

Distributed deep learning operation and maintenance system

Tourist hotspot event detection method and system