<img height="1" width="1" style="display:none;" alt="" src="https://px.ads.linkedin.com/collect/?pid=3740828&amp;fmt=gif">

Oct 27, 2022 9:00:00 AM | Shoplazza News 2022 AWS Summit China: Shoplazza Shares AI Innovations & Practices

Shoplazza CTO Bing Xia shares advances in AI-powered recommendations systems at the 2022 AWS Summit China.

October 13-14, 2022, AWS Summit China - "Build with Freedom, Explore the Infinite". At the event, tech experts and enterprises shared their global perspectives on cloud computing, plus innovative practices and insights. Shoplazza CTO, Bing Xia, shared advances in the platform's AI-powered recommendation system for cross-border ecommerce.

Bing Xia at the 2022 AWS Summit China.

Bing at the 2022 AWS Summit China

The summit, hosted by Amazon Web Services (AWS), is China's largest annual technology conference and a benchmark event in global cloud computing.

As a global leader in shopping cart SaaS, Shoplazza was invited to share its exceptional practices in advanced product features and technology.

Shoplazza provides 360,000+ merchants worldwide with various enterprise-level solutions, including curated ecommerce store themes, automated order management and store operations, multi-channel marketing campaigns, plus intuitive and secure payment processes.

In recent years, branding has played an increasingly important role in ecommerce. More and more cross-border merchants have adopted the direct-to-consumer (DTC) model and are moving towards more refined operations and management to grow their businesses. One of the crucial features implemented in their online stores to that end is the "AI-powered recommendation" system.

"On a global scale, AI-powered recommendation is an essential technology for improving sales revenue and customers' shopping experience," Bing explained.

"But it's tough for small to medium-sized businesses (SMBs) to implement AI solutions designed for large companies," he says. "SMB merchants have to put extra effort and time into manually optimizing product recommendations, which undermines operational efficiency."

Keeping customers happy with their recommendations can also be a challenge.

"That's why we aim to provide an AI-powered tech solution for businesses of all sizes – our recommendations system can lower costs and improve efficiency for any business."

Bing points out that Shoplazza's AI and machine learning system has five stages:

(1) data crawling,

(2) triggering,

(3) pre-ranking,

(4) ranking, and

(5) re-ranking.

Shoplazza is highly focused on data processing to ensure that every step of the process runs efficiently and smoothly. Open-source software libraries like TensorFlow cannot train AI quickly enough due to the massive amount of data required to train AI to interpret transaction amounts and user access. Additionally, distributed machine learning on Spark has issues like slow training and high cost.

To address these issues, Shoplazza uses the data flow and computing power of Amazon EMR and Amazon SageMaker to combine sampling, feature training, and estimation for Shoplazza's recommendation system data flow. This combined procedure enables more efficient data processing and machine learning capabilities. 

Bing briefed attendees on the detailed data flow implementation for AI-powered recommendations with SageMaker.

"Model training is performed in the SageMaker platform using generated samples by deploying real-time reasoning. When a user requests recommendations, AI prediction can splice user features and product features in real-time, display the feature extraction plugin, and generate samples, while requesting scoring from SageMaker and displaying the sorted results on the webpage."

With Amazon EMR and Amazon SageMaker, Shoplazza has been able to ease the management of big data and machine learning infrastructure, combining data analysis and machine learning for cost-efficient results. 

"It only takes 2 hours to process 300 GB of training data," Bing pointed out.

"We reduced the time requirement by about 10 hours and improved the product recommendation system performance by 600%. We continue to implement and develop AI technology to serve our merchants better and help them achieve global success."

Bing concludes, "AI and machine learning are the most transformative technologies in the world today. They significantly enhance customer experience and enterprise productivity. Shoplazza will continue to work with AWS to integrate machine learning recommendations into product search, content processing, merchant selection, risk control, and other ecommerce business operations. Tech innovation will solve merchant store operation struggles."

As a globally leading ecommerce SaaS platform, Shoplazza is committed to continuously providing innovative solutions and sharing successful practices for market growth.

Shoplazza remains "open to more" through our Partner Program. We continue to optimize our partner ecosystem and work with AWS to help cross-border merchants "start anywhere" with the help of our all-in-one solution for going global, encompassing the latest technology, product innovation, and data strategy.

Sign up to Shoplazza today to start your business.

Written By: Shoplazza

Shoplazza is a cloud-based eCommerce platform for online retailers.