亚信科技

 A few days ago, a number of industrial annual lists were announced by “Data Yuan.cn”, an authoritative leading media focused on the field of big data. AsiaInfo was successfully shortlisted in three important lists, with several products enrolled into “Chinese Data Intelligence Industry Graph”.

The industrial annual lists of “Data Yuan.cn” aim to discover companies, products and technologies worthy of attention in the big data field in 2020 from a professional perspective. Through two months of preliminary examination, re-examination and expert evaluation, AsiaInfo stood out from hundreds of participating companies and was shortlisted in the list of “Company with Innovative Service in 2020 Big Data Industry”, its data exploration and analysis platform AISWare DataDiscovery was awarded the “Annual Innovative Service Product”, and the anomaly detection technology for key indicators of intelligent operations was shortlisted in the “Innovative Technology Breakthrough List”. The following are AsiaInfo’s technologies, products and cases referred in the lists.

AISWare AIOps: Anomaly detection technology for key indicators makes more accurate, smarter and more efficient operations

In the field of operations, the monitoring and warning of key indicators are methods commonly used to discover problems. Traditional methods of anomaly warning are mostly centered on static thresholds, which requires manual setting of thresholds for different types of indicators and instances based on experience. Therefore, the configuration and maintenance workloads are too large to achieve network-wide coverage. At the same time, as the static threshold is less flexible, the threshold is prone to fail when the business changes, and is not sensitive to local anomalies in the cycle.

AsiaInfo’s intelligent operations products focus on the pain points of the basic operations system, carry out AI algorithm innovation and conduct componentized encapsulation of capabilities by centering the entire process of fault discovery, diagnosis, disposal, prevention, as well as typical scenarios and requirements such as intelligent Q&A and resource optimization, in order to empower and add intelligence to the operations system. Aiming at the above monitoring and warning scenarios, the key indicator anomaly detection technology developed by AsiaInfo on its own is characterized by excellent accuracy, timeliness and adaptability to data with different characteristics, and can be flexibly adapted to the key indicator monitoring in multiple types of complex operations scenarios.

This technology automatically recognizes data fluctuation characteristics by constructing an unsupervised anomaly detection model, and selects differentiated algorithms for model training and real-time anomaly diagnosis for data with different fluctuation characteristics. At the same time, it aggregates and reports anomaly warnings in combination with business specifications. Operational data shows that this technology’s fault recall ration can reach 99%, and the precision ration can reach more than 90%. Compared with open source algorithms such as LSTM, the resource consumption of this technology can be reduced by about 50%.

At present, this technology has been widely used in the intelligent operations practices of major domestic telecom operators, and is suitable for industries such as power, radio and television, finance and energy. AISWare AIOps, AsiaInfo’s intelligent operations product which uses this technology, has won several industry awards, including the runner-up of Third International AIOps Challenge 2020, the “Best Catalyst” of TMF Asia Summit, etc., and the cases such as “Network Operations Anomaly Detection based on Dynamic Threshold” and “Cross-domain Intelligent Alarm Root Cause Analysis” have been included in Smart Autonomous Network Case Report of GSMA.

AISWare DataDiscovery: Data exploration and analysis platform digs into the value of data to contribute to the efficient operation of companies

At present, the demand for high-quality tools bears the brunt in the big data self-help analysis and data value digging of companies. AISWare DataDiscovery, AsiaInfo’s data exploration and analysis platform, arose to the occasion. It can help companies respond to many demands such as precision marketing, data asset management, business analysis and decision support, system operations diagnosis.

The product is a tool for agile and open data exploration and visual analysis, including three sub-products of “self-help fetch”, “data factory” and “indicator exploration”. It provides self-help, visualized and integrated data exploration and analysis capabilities, with many advantages such as inclusiveness, self-help, practicality, safety, and open ability. Relying on this platform, business analysts can quickly understand the nature of data, carry out drag-and-drop data analysis and design in a convenient manner, and build a bridge between questions and answers.

In terms of functions, the product provides multi-source data access and integration capabilities, and achieves flexible dashboard data definition through the WYSIWYG” data visualization configuration. With various data expression forms, the abundant visualization components can meet the data presentation in different scenarios (such as: indicator card, Sankey diagram, box plot, scatter plot, map, etc.); data analysis not only supports routine analysis (such as: cycle comparison, calculation of indicators, grade-statistics, etc.), but supports complex scientific calculations, and integrates multiple calculation methods internally (such as: Sarima prediction, normalized analysis, correlation analysis, etc.); it supports consolidating analysis results into online analysis reports (the format similar to PPT), supports online play, and generates editable PPT offline; it provides self-help fetching capabilities, and data screening, combination and secondary calculation to meet fetching scenarios (such as target customer marketing, etc.); it supports real-time data refresh, making the data analysis process more dynamic to be suitable for scenarios such as real-time indicator monitoring, operations support, etc.; it supports multi-terminal adaptation (such as : large screen, mobile terminal), and can be posted to social platforms such as DingTalk and WeCom; it provides a variety of security mechanisms to ensure data security, which can effectively prevent unauthorized inquiry and leakage of sensitive data. The product has been applied in the fields of communication, fire protection, government affairs, finance, etc.

AsiaInfo helps industry companies reduce costs and increase efficiency through innovative big data services

Called as the “blood of the digital economy”, data is an important information asset characterized by massive amount, high growth and diversification. Various industries are committed to the development and utilization of data assets to provide an inexhaustible supply of “fuel” for efficient production, operation and management of companies and institutions. AsiaInfo comprehensively utilizes innovative big data services to help all industries reduce costs and increase efficiency.

A province in South China initiated a provincial-level highway big data platform construction project during the cancellation of provincial boundary toll stations of highways, hoping to improve the centralized control of provincial central data, enhance the timeliness and integrity of data intercommunication with national and ministerial-level networking centers, reduce the occurrence of toll evasion of highway vehicles, improve the IT information level and business-oriented support capabilities of provincial highway systems, and increase the efficiency of highway management and audit staff.

Faced with the realities of many highway systems, complex data formats, and the alienation of systems and businesses of the province, AsiaInfo utilized big data and AI technologies and products to build a data aggregation system, data sharing center, audit and credit management system, risk management system and business data analysis system to realize the unified aggregation, analysis and service of the province’s highway data, so that the provincial big data platform was equipped with the ability of whole-process and refined analysis and management of highway business, which provided strong support for a refined management and operation of highway.

Launched for less than one month, the platform has completed a unified audit of all historical abnormal data in the past six months, achieving the goal of a full range of audit and creating a record of checking and successfully recovering tens of millions of evaded tolls within one month. It effectively safeguards the rights and interests of investors of highway construction, prevents the loss of state-owned assets, and effectively suppresses and cracks down on toll evasion.

In the process of serving large companies for more than 20 years, AsiaInfo has accumulated abundant experience in big data product development and business practice, and has formed a relatively complete AISWare BigData big data product system (the data exploration and analysis platform AISWare DataDiscovery is one of them), which can be widely used in data mining, analysis, processing, governance and application in various industries to provide companies with stronger data insight, decision-making power and process optimization capabilities.