Secure architectures and methods for improving the security of mobile devices are disclosed. Also disclosed are apparatuses and methods to detect and mitigate fraud in device-assisted services implementations.

Unauthorized and fraudulent use of a cloud computing system may be reduced or mitigated using a multi-threshold based method to identify fraudulent subscribers of the cloud. The multi-threshold based method may assign a fraud threshold to each resource of the cloud. The fraud thresholds of the multi-threshold based method may be adjusted based on one or more characteristics associated with one or more of the plurality of resources in the cloud. The one or more characteristics may include a capacity percentage associated with the plurality of resources, fraud distribution among the plurality of resources, cost of operation associated with the plurality of resources, anticipated or actual subscriber growth rate associated with the plurality of resources and/or anticipated or actual subscriber fraud risk associated with the plurality of resources.

Detecting fraudulent activity can be a complex, manual process. In this paper, we adapt statistical properties of count data in a novel algorithm to uncover records exhibiting high risk for fraud. Our method identifies shelves, partitioning data under the counts using a Student's t-distribution. We apply this methodology on a univariate dataset including cumulative results from phone calls to a customer service center. Additionally, we extend this technique to multivariate data, illustrating that the same method is applicable to both univariate and multivariate data.

Unsolicited electronic communications such as robocalls and person-initiated solicitation calls are reduced by imposing tolls for completion of the connections to the called parties, and refunding the tolls to the entities indicated by the electronic communications as the calling parties. In this way, a dishonest originator of a spoofed call bears the cost of the toll, and the toll is not refunded to the dishonest originator. On the other hand, the toll collected from an honest originator of a non-spoofed call is refunded to the honest originator, making the toll transparent to the honest originator and avoiding annoyance of the honest caller caused by the toll. Unsolicited calls may be subjected to filtering, particularly filtering based on the indications of the origins of the calls.

The present specification provides a blockchain-based data verification system and method, a computing device, and a storage medium. The blockchain-based data verification system includes: a first verification system, a second verification system, and a first blockchain node and a second blockchain node on a blockchain network; the first verification system is configured to: collect first service data based on a predetermined condition and generate a first verification file including the first service data, and send a verification request to the second verification system; the second verification system is configured to: receive the verification request, verify the first service data with second service data in a local database of the second verification system, generate and send a verification result notification to the first verification system, and send successfully verified second service data in the local database to the first blockchain node on the blockchain network; wherein the first verification system is further configured to: receive the verification result notification, and send successfully verified first service data to the second blockchain node on the blockchain network.

An information verification method and a related device. The method includes: receiving, by a first profile server, a remote profile management request sent by a remote profile management server, where the remote profile management request includes a first profile owner identifier; obtaining a second profile owner identifier in profile information; determining whether the first profile owner identifier is the same as the second profile owner identifier; and sending a remote profile management command to user equipment when the first profile owner identifier is the same as the second profile owner identifier. In this way, a loss caused to an operator or a profile owner due to a user's access to an invalid network or random profile unlocking is avoided, network security is improved, and waste of network resources is reduced.

Telecommunication services systems and methods are disclosed. The telecommunication services systems and methods preferably enable a buyer to certify and activate a used wireless telecommunications device. The systems and methods may also be employed to provide the buyer with specialized pricing options and other customized services.

Methods and systems are presented for generating a machine learning model using enhanced gradient boosting techniques. The machine learning model is configured to receive inputs corresponding to a set of features and to produce an output based on the inputs. The machine learning model includes multiple layers, wherein each layer includes multiple models. To generate the machine learning model, multiple models are built and trained in parallel for each layer of the machine learning model. The multiple models use different subsets of features to produce corresponding output values. After a layer in built and trained, a collective error may be determined for the layer based on the output values from the different models in the layer. An additional layer of models may be added to the machine learning model to reduce the collective error of a previous layer.

Outgoing communications from a user device are monitored following a notification that an application is installed on a user device. When a number of the outgoing communications to a premium-rate number during a predetermined period of time exceeds a corresponding quantity threshold, the pattern of the outgoing communications may be ascertained to be anomalous. A user device is directed to present a prompt that requests an input as to whether the outgoing communications are authorized. In response to a first input that the outgoing communications are unauthorized, additional outgoing communications from the user device to the premium-rate number are blocked for a predesignated amount of time. Further, the first input is stored as a corresponding vote that the application is malicious. In response to a second input that the outgoing communications are authorized, the second input is stored as a corresponding vote that the application is non-malicious.

A Toll-Free Management Platform (TFMP) includes methods and systems for number administration, customer administration, call management services, texting services and text registry, and a smart services registry. The TFMP may allow users to receive through a web-based interface, an activate request from a user, wherein the request includes at least a customer record template reference and an indication of when to active a toll-free telecommunications number associated with the request. The request may include at least one IP address datum associated with the toll-free telecommunications number. A user interface may allow activating a toll-free number, and users may access the TFMP to create and access existing templates of toll-free call routing templates, and utilize a routing tree engine to create customized call routing trees for the toll-free numbers of interest to the user.