APEX Technologies
Published in

APEX Technologies

APEX Technologies uses blockchain technology combined with AI to achieve cross-enterprise data exchange and insight

Author: Blockchain Product Manager Wang Yifan Jude. Responsible for the design and engineering realization of APEX Federated Learning products, and has in-depth research and insight on cutting-edge technologies such as machine learning and blockchain.

As Apple released the iPhone in 2009, the smartphone revolutionized the mobile Internet, making it easy for companies to collect consumer data. Companies have access to massive amounts of consumer data, and these consumer data be utilized to accelerate the company’s development.

Simultaneously, current artificial intelligence relies on massive consumer data. In general: The more consumer data, the better performing artificial intelligence model.

To this end, the industry has developed the concept product CDP (Consumer Data Platform), to standardize and unify data within the enterprise and to collect more potentially valuable data within the enterprise.

However, some businesses rely heavily on third-party data for their business, and many companies need third-party data to improve the performance of their own AI models.

Because data is related to the pace of enterprise development, there are numerous cases of data abuse in the industry

  • In June 2018, Facebook provided business partners including smartphone, tablet and other device manufacturers with in-depth access to users’ personal information, allowing some companies to effectively access users’ privacy settings. Cambridge Analytica used this to manipulate the 2016 US election.
  • In September 2019, many domestic third-party data companies in China were investigated. A financial data company illegally made more than 1 billion yuan in profits!

The chaotic area of data abuse will eventually be ended by regulation, and the entire industry will become more and more standardized.

  • On May 25, 2018, the European Union issued the “General Data Protection Regulation”.
  • In April 2019, the Central Bank issued the “People’s Bank of China 2019 Regulations Work Plan”, which includes “Personal Financial Information (Data) Protection Trial Measures”.
  • May 25, 2020, “Li Zhanshu: Personal Information Protection Law will be enacted”

To this end, we can’t help but think about ways to protect consumer data while allowing companies to share data to improve the performance of their respective AI models.

We integrated technologies such as blockchain, cryptography, and machine learning, and launched APEX Federated Learning.

APEX Federated Learning

APEX Federated Learning works out-of-the-box, and combined with APEX NEXUS it enables users to independently build a marketing focused CDP based on private domain data, helping the enterprise achieve new insights, cultivate and transform potential customers, retain old customers, improve communication and marketing, etc.

During the important data collection stage, optimized through APEX Federated Learning, enterprises using their CDP can quickly learn similar characteristics of data, helping them quickly identify groups of target data.

At the stage of cultivating potential customers, APEX Federated Learning enables enterprise users to quickly obtain more dimensional information about users from data with very few user attributes. This allows enterprises to swiftly determine whether the potential customers are high-value and high-intention users, while avoiding the security and legal risks of tripartite data transactions.

APEX Federated Learning is based on cryptography and blockchain technology, which enables multiple companies to decentralize the construction of AI models and services. No one company can obtain or crack the data of another company, and the system keeps even the distribution and upper and lower bounds of the data secure.

The necessity of APEX Federated Learning

APEX Federated Learning complements CDP and NEXUS, aiming to break down the data silos between enterprises, improving enterprise AI efficiency to accelerate company development.

As an example, to outline the APEX Federated Learning modeling process, consider these two companies: Large retail company R and car company C.

Car company C hopes to build an AI model to predict whether a consumer is a potential customer. Car company C already has some consumer data, such as shopping guide consulting data for consumers and corporate account managers, but these data are not sufficient to create an excellent AI prediction model.

Car company C queries the potential data service companies in the APEX Federated Learning server. Car company C finds that there are many intersections between its existing consumers and retail company R. Because of the RSA intersection, the two companies can only know which consumers are the intersection customers of the two parties in this process, and the other customers have no information about the other party.

At this time, auto company C initiates the APEX Federated Learning modeling task, and retail company R provides data services. In this process, the data provided by auto company C to retail company R is completely aggregated data, homomorphically encrypted. The entire process is completely decentralized, and key logs are recorded on the blockchain. When the modeling is completed, the prediction task of car company C needs the assistance of retail company R. In this process, car company C achieved improved performance of their model. Retail company R provided data services for car company C. For this service car company C will be charged a certain fee. And for the consumer data used in the modeling and forecasting, consumers will receive certain consumption discounts. Of course, retail enterprise R needs to obtain consumer authorization to provide forecasting services.

Since APEX Technologies already has many enterprise customers, existing customers can be directly converted into members of the APEX Federated Learning Alliance, and enterprise members of the entire alliance can be both data requesters and data providers. When the APEX Federated Learning ecosystem further flourishes, the AFL Alliance can further develop a blockchain based digital credit business to further enhance consumer loyalty within the alliance, making the APEX Federated Learning ecology a closed-loop system.

APEX Federated Learning removes third-party data service companies and decentralizes the cost of obtaining external data for enterprises, which can maximize the interests of both consumers and enterprises to a certain extent.

Data requesting enterprises can also view important information such as model stability, input-output ratio and other important information obtained through APEX Federated Learning, and set up follow-up marketing behavior to monitor the model status in real time.

Due to the privacy and security of APEX Federated Learning, APEX Federated Learning is more secure and compliant than third-party big data companies. Compliant with the regulatory requirements of the government, it can reduce the operational risks of enterprises.

APEX Federated Learning’s data privacy and security makes it easier to gain consumer trust and enhance the company’s brand image and value, which will benefit the company’s long-term development.

For companies, from a strategic perspective, those that are among the first to join APEX Federated Learning can gain a first mover advantage and can take the lead in mining the value of data within the alliance.

APEX Federated Learning — Product Advantage

APEX Federated Learning is superior to traditional third-party data companies in many ways

  • Based on cryptography technology, APEX Federated Learning does not disclose the original data. Even the distribution of the original data, including the upper and lower bounds, will not be disclosed
  • APEX Federated Learning performance is excellent, and compared with the direct modeling of original merged data, the performance is lossless
  • APEX Federated Learning supports multiple AI algorithms, and the encryption algorithm does not limit the freedom of machine learning
  • APEX Federated Learning learns from the idea of ​​layer two in blockchain technology, and records necessary log information on the chain to prevent any potential malicious enterprises from doing harm
  • APEX Federated Learning uses the blockchain as its center. The entire modeling is completely decentralized, and no subject can obtain or crack the data held by other subjects. Completely removing third-party centralized services and reducing data acquisition costs.

Originally published at: https://mp.weixin.qq.com/s/F4V7NvFzh9tsaJtfwrtIEQ

About APEX Technologies

APEX Technologies provides an enterprise customer data technology solution that helps marketers connect, unify, analyze, and activate their customer data.




APEX Technologies provides an enterprise customer data technology solution that helps marketers connect, unify, analyze, and activate their customer data.

Recommended from Medium

Rise of Artificial Professionals


Please don’t rely too heavily on Grammarly. It is not always correct. Case in point:

Artificial intelligence

Proposal for GANs AI-based Texture Generation for 3D Developers (Abstract)

Kickstart to career in AI with Meta’s RAISE Program

Analyzing and Improving a Watson Assistant Solution Part 1: Analytics Personas and Existing…

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store


Blockchain Powering the Next Generation of Consumer Applications

More from Medium

Streaming fatigue part 2

Image of wolf

Our Investment in Edge Delta

Solving metadata mutability restrictions in ERC721 Standard for real World Aplications.

Joseph Bondaryk joins Toggle Robotics as CTO

Engineers at Toggle in the R&D facility