:::

【2021 Solutions】 Action Bagel makes AI as simple and efficient as Excel to improve data analysis capabilities

What is AutoML (automated machine learning) and how is it different from ML (traditional data analysis)? It needs further clarification first.

Traditional machine learning must go through data cleaning, data pre-processing, feature engineering, feature selection, algorithm selection, model establishment, model training, parameter adjustment, and then evaluation results to produce model applications. During the process, if there is a problem with the parameters, the algorithm must be re-selected, the model must be re-established, etc., and the process must be repeated hundreds of times. If new information becomes available, all steps must be repeated. Through automated machine learning, the output process of model application only needs to go through the automation of four major steps: data cleaning, feature engineering, data modeling and model evaluation to achieve model application. Even if new data needs to be collected, it can be achieved through Automated machine learning is achieved, saving time and effort.

ML and AutoML Comparison source: Action Bagels Ltd

▲Comparison between ML and AutoML Source: Action Bagel Co., Ltd.

AutoML is a program that can automate the time-consuming and repetitive work in machine learning model development. This allows small and medium-sized enterprises that relatively lack AI talents to create their own customized machine learning models. In recent years, major international companies have rushed into this market, including Cloud AutoML released by Google in 2018, and AutoPilot launched by cloud computing leader AWS in 2019. AutoML has become a standard feature of mainstream learning services, from web-based interfaces to free Program development and workflow visual management, etc., service development is becoming more and more diversified.

MoBagel is a professional team composed of top data scientists, engineers, and product project managers. The team members come from prestigious universities around the world, including Stanford, Berkeley, Oxford, and National Taiwan University in the United States. They also have experience in Selected to participate in Silicon Valley's well-known accelerator 500 Startup, selected to participate in Japan's SoftBank Innovation Program, and also won a name in Nokia's Open Innovation Challenge.

Mobile Bagel Decanter AI platform shortens the analysis project from two months to two days

Mobile specializes in data science and machine learning technology. In 2016, it developed the automated machine learning analysis tool Decanter AI. So far, it has helped more than 100 companies introduce AI into important decisions, and the analysis project has been shortened from two months to Two days. The fields served include retail, telecommunications, manufacturing, finance and other industries.

Lin Yushen, deputy general manager of Action Bagel Co., Ltd., said that Decanter AI makes AI as simple and efficient as Excel, which can improve enterprise data analysis productivity. Users do not need to have in-depth professional knowledge and experience. Through a simple super-operating interface, they can perform automated machine learning for data analysis and prediction.

There are three simple steps to use Decanter AI: Step 1. Organize the data into csv format; Step 2. Upload to DecanterAI to set prediction goals; Step 3. Decanter AI automatically models and obtains prediction results. The deployment method can be in the public cloud or in the private cloud of the enterprise. After the internal data is uploaded, it can be modeled and used.

DecanterAI uses three steps, Simple and convenient

▲DecanterAI uses three steps, simple and convenient

The advantage of AutoML is that it can automatically train a large number of models, adjust parameters, produce the best model, and quickly deploy and import it. After the new coronavirus (COVID-19) epidemic, all walks of life are facing new market changes and must Transform digitally with fast and convenient digital tools.

In recent years, Action Bagel has continued to promote the optimization of the DecanterAI platform and establish industrial data modeling and analysis capabilities, and has produced substantial results. For example, Chunghwa Telecom uses its platform to conduct blind tests on code-carrying customers and perform data analysis to effectively reduce user churn rates and improve customer retention rates. As a leading domestic food manufacturer, due to the expiration date of drinks and the production and sales of the cold chain, it must be fully integrated to reduce inventory and loss problems. After importing the DecanterAI platform, in addition to accurately predicting market demand, it can also accurately predict market demand based on expiration date data. Analyzing production and distribution quantities can also help reduce warehousing and logistics costs.

AutoML industry has diverse and extensive applications and great potential for future development

Action Bagel believes that AutoML has a wide range of industrial applications, including employee turnover prediction, production demand prediction and revenue performance prediction that are troubled by the manufacturing industry; store passenger flow prediction, product replenishment prediction, membership prediction in the smart retail industry Promotional forecasting; customer churn forecasting and potential customer list forecasting in the telecommunications industry; accurate financial marketing, credit card fraud detection and insurance application quick review in the financial industry; and even real estate price forecasting, power outage disaster forecasting, etc. are all helpful. To solve the operating difficulties of the industry and create new business models.

AutoML industrial applications are diverse. Covering manufacturing, retail, finance and other industries. Source: Action Bagel Co., Ltd.

▲AutoML has diverse industrial applications, covering manufacturing, retail, finance and other industries. Source: Action Bagel Co., Ltd.

How much time and preparation does it take to import AutoML? Lin Yushen said that in actual practice, the introduction process of automated machine learning enterprises includes four major stages:

1. Preparation period: Collaborate with enterprises to discuss business pain points, help define analysis propositions, and provide professional data science advice and optimal solutions, lasting about two weeks.

2. Verification period: Use a small-scale pilot project to quickly verify the analysis results to ensure proposition setting, data quality, analysis process, prediction technology, etc., as the basis for subsequent practical application and amplification. It takes three weeks.

3. Introduction period: Support cloud or local product deployment according to enterprise needs. Provide operation and maintenance teaching, Help Center, data analysis consulting, corporate training courses and other product introduction services, which will take more than one month.

4. Application period: Analysis/data teams can execute various AI projects through the product's common interface and implement them quickly. The prediction engine can be connected through the API to develop application modules according to practical scenarios. This is the final stage of application and is time-consuming and can take up to several months.

However, Action Bagel conducts a system integration project process with its SI partners. The SI partners discuss business propositions and provide data sets, and then conduct data health checks and Baseline models. Based on this, Action Bagel provides data diagnosis reports. After confirming the pilot project proposition and producing a demand planning document, the project execution phase begins, with model establishment, optimization and analysis reports provided. System integration with SI industry players, on the one hand, optimizes module development, and on the other hand, uses APIs to connect data sources and output prediction results, import them into the enterprise's field, and effectively solve the propositions faced by enterprises in digital transformation.

Looking forward to the future, Decanter AI platform will continue to develop various AI innovative application services, and cooperate with the upstream, midstream and downstream industries such as enterprise resource planning (ERP), customer relationship management (CRM), business analysis (BI) and e-commerce platforms ( EC) and other partners maximize the benefits of the ecosystem through co-creation, sharing and altruism.

(This article is derived from the selected content of "AI Engineering Online Small Gathering")

「Translated content is generated by ChatGPT and is for reference only. Translation date:2024-05-19」

Recommend Cases

【解決方案】7毫秒內分離人聲 洞見未來科技協助聽損者「聽說更簡單」
Voice Separation in 7 Milliseconds: RelaJet's Future Technology Makes 'Hearing and Speaking Easier' for the Hearing Impaired

One rainy Thursday afternoon near Taipei Arena, the Taipei Experience Center of RelaJet was fully booked with appointments from people with hearing loss eager to try hearing aids made with a voice separation engine For the hearing impaired, having affordable, lightweight, and effective noise-reducing hearing aids is truly a blessing 'We hope to help users in need to hear the world's wonders again' This empathetic expectation by RelaJet's founder and CEO Po-Ju Chen, who is also hearing impaired, illustrates his understanding of the needs of the hearing impaired He hopes that RelaJet's unique voice amplification hearing aid technology will benefit many more people Affordable hearing aids benefit many with hearing loss Founded in 2018 by Po-Ju Chen and his brother Yu-Ren Chen, RelaJet developed a multi-voice separation engine paired with Qualcomm's Bluetooth audio platform, drastically reducing the price of imported hearing aids, typically costing 80,000-100,000 NT dollars, to just under 10,000 NT dollars They aim to develop affordable goods with excellent noise-cancelling capabilities that wirelessly connect to smartphones In its first two years, the company primarily developed the multi-voice separation engine, which significantly improved the noise reduction quality Once equipped with Qualcomm Bluetooth earphone chips, the audio processing time is drastically short, at about 7 milliseconds to enhance main voice projection and reduce ambient noise, less than half the time required by traditional medical standard of 16 milliseconds for hearing aids, nearly 'zero-delay' 洞見未來科技推出平價助輔聽器,大大嘉惠聽損者 Yu-Ren Chen explains that the primary use of Qualcomm chips for edge computing, along with a streamlined algorithm, achieves extremely low latency and better noise reduction The hearing aids can cover 18 channels, whereas traditional hearing aids cover 4-48 channels In the future, RelaJet will progressively increase the number of channels According to statistics, there are 470 million people globally with hearing disabilities, with a 30 average device use rate in developed countries, with the highest in Western countries Taiwan has nearly 15 million people with disabling hearing loss, of which the middle-aged and elderly make up 30, yet the device use rate is only about 10, which is quite low Yu-Ren Chen further analyzes that the low device usage rate is due to two reasons firstly, the high average selling price of international big brands ranges from 80,000 to 200,000 NT dollars with a three-year usability period, which deters many due to the high cost and maintenance Secondly, in noisy environments, the noise is also amplified which does not necessarily ensure clarity for the users, and the sound parameters can't be adjusted in real-time, making it inconvenient to frequently visit stores for tuning Thirdly, most models cannot connect to smartphones, making it inconvenient for the hearing impaired to take phone calls Utilizing Qualcomm Bluetooth chips for rapid product development In light of this, Po-Ju Chen, formerly a semiconductor engineer at MediaTek, leads the technical development, while Yu-Ren Chen, with a legal background, manages the operations Their seamless collaboration, along with their team employing AI algorithms and chip integration, learns from thousands of hours of audio files in databases through neural networks and deep learning technologies to develop low-latency, high-noise-reduction voice amplification technologies for hearing aids In 2019, this sound processing technology was integrated into Qualcomm Bluetooth chips, winning first place in the Qualcomm Taiwan Startup Competition and becoming a partner in Qualcomm's Global Expansion Program, significantly boosting product development pace In 2021, they launched their own Otoadd series of hearingenhancement products in Taiwan, which received both market favor and positive reviews from many with hearing loss Based on different consumer needs, various product designs are available According to Yu-Ren Chen, the Otoadd wireless earphones with hearing enhancement functions, model N1, are entry-level neckband style priced at 9,500 NT per pair Users can wear the hearing aid while taking calls, and control noise reduction strength and volume through a mobile app They plan to develop accessories in the future tailored to the needs of older adults Besides being available for trials at experience centers in Taipei and Kaohsiung, this hearing aid is also sold through PChome, Taiwan Mobile's myfone, and Elder Age networks, among other channels Another model intended for individuals with mild to severe hearing loss is the Classic R hearing aid, which received the Japanese Good Design Award in 2021 Since its market debut last year, it has attracted those with congenital hearing loss, with users noting improved clarity in noisy environments and appreciating the convenience of Bluetooth connectivity for calls and watching videos This product is anticipated to be exported to international markets in the latter half of this year Additionally, a hearing aid product combining Bluetooth functionality, set to launch in June this year, is sized like typical Bluetooth earphones, targeting visually conscious consumers with hearing loss Its small size and attractive wireless earphone design allow for phone calls, and if approved by the Ministry of Health and Welfare, eligible users can apply for government subsidies RelaJet to expand into overseas markets, using the USA as a beachhead An interesting question arises due to the pandemic everyone must wear masks which impedes lip-reading How does this affect those with hearing loss Yu-Ren Chen indicates that this situation highlights RelaJet's advantages As each person with hearing loss has different levels of hearing ability, hearing aids can only augment to an appropriate volume, assisting users to hear about 60-70 content, with the remainder relying on lip reading and gestures During the pandemic, as everyone wears masks, masks also muffle sounds, but RelaJet's voice separation engine can correct and strengthen the separation, making it easier for those with hearing loss to recognize voices Besides the Taiwan market, RelaJet's next stage will be expanding into overseas markets, expecting to obtain ISO 13485 medical device quality management system certification and US medical device approval in 2022 They plan to enter the US market, either under their own brand or through OEM arrangements Apart from the Taiwan market, RelaJet will also enter the US market in the next phase for hearing aids「Translated content is generated by ChatGPT and is for reference only Translation date:2024-05-19」

【解決方案】AI電眼取代人眼 慧演智能運用AI幫製造業做品管
Using AI vision to replace human vision, Claireye Intelligence uses AI to help the manufacturing industry with quality control

In response to customer demand on a wide variety of products in small quantities in the manufacturing industry, there is an urgent need to find AI solutions from the cloud to terminals Claireye Intelligence provides a solution that integrates software and hardware - BailAI image inspection solution to assist traditional manufacturing industries in improving process efficiency and product quality, thereby achieving the initial goal of transformation After the government declared 2017 to be Taiwan's "First Year of AI," AI startups have sprung up in Taiwan Established in 2018, Claireye Intelligence targets smart manufacturing and provides a platform for AI image analysis and process optimization, using the power of deep learning to detect product defects and abnormalities in the assembly process It assists companies in building infrastructure from terminals to the cloud, which enables automated monitoring of factory production to improve process efficiency and quality Focusing on AI image inspection based on its familiarity with the production line quality control process Shirley Liu, founder and CEO of Claireye Intelligence, is a young entrepreneur She entered the manufacturing industry after graduating from college and held a quality control position in the plastic injection process of hard disk parts "She was already on the production line at the time, and is familiar with the production line process of production machinery" She later switched career paths to marketing and planning, and then worked as an AI product manager When the time came, Shirley Liu decided to start a business, focusing on AI image recognition in the manufacturing industry "The difficulty for enterprises is the lack of an AI development team Even if an enterprise has an AI team, development projects will take a lot of time, at least 6-12 months" said Shirley Liu, who is well versed in the market's pain points The problem that needs to be solved by platforms is to provide services that allow traditional manufacturing industries to build their own AI models without needing employees with a programming background, and to remotely assist production lines with troubleshooting and subsequent system maintenance, helping companies save development time and labor costs BailAI image inspection platform usage scenarios Facing the large number of competitors that provide AI image recognition in the market, what are the technical advantages of Claireye Intelligence Shirley Liu said that many companies currently have AOI equipment, but the bottleneck in the application of AOI is that it can only be used for defect inspection in fast production of large quantities, and parameters need to be adjusted after each inspection or production Based on her understanding of the industry, most SMEs are limited by their financial resources due to AOI equipment often costing over NT1 million, but they also want to use automated inspection This is where Claireye Intelligence comes in Shirley Liu went on to say that it is impossible for traditional manufacturing industries to maintain a technical team that includes AI engineers, data engineers, cloud architects, and terminal architects Claireye Intelligence specializes in software and hardware integration Enterprises can use the BailAI image inspection platform to easily solve inspection problems on the production line In other words, customers only need to provide images or samples for Claireye Intelligence to carry out model training, model deployment, and system integration, and they can easily use AI technology to optimize and monitor production line processes Participated in the AI New Talent Selection and achieved a recognition rate of over 90 in assembly behavioral image recognition For example, a certain connector manufacturer only has 1-2 AI engineers in its technical team The main problem that needs to be solved is that most operators are on the production line, while quality control and senior managers are not on site, and the company wants to understand the actual situation of the production line through remote monitoring Claireye Intelligence uses industrial cameras to capture production line images, and transmits AI image analysis to the remote end Supervisors and quality control personnel can observe if there are any errors in the production line assembly, such as whether the connectors and lines are connected properly, through the monitor Claireye Intelligence's AI image inspection operates on Microsoft's Azure cloud platform, and also utilizes terminal equipment, such as NVIDIA's edge computing equipment placed around the inspection station, to assist traditional manufacturing industries with improving production line efficiency and detecting problems early through an integrated solution from the cloud to terminals Claireye Intelligencersquos customers currently include aviation, electronic peripherals, connectors, and metal industries Assembly process solution for human behavior recognition in assembly lines achieves an accuracy of over 90 In order to demonstrate the depth of technology, Claireye Intelligence participated in the 2021 AI New Talents Selection of the Industrial Development Bureau, Ministry of Economic Affairs, and provided Lite-On Technology with the "assembly process solution for human behavior recognition in assembly lines" The solution determines effective working hours and ineffective working hours of operators on the production line through cameras and AI image recognition It recognizes hand posture and position through images to determine the operator's assembly behavior, achieving an accuracy of over 90 Shirley Liu added that the assembly process of electronic components is complex, mostly carried out manually, and cannot be replaced by robotic arms Claireye Intelligence used cameras to film the assembly process of operators at Lite-On's assembly station The algorithm is then trained and corrected based on the video, and the final trained model can directly determine whether there are any errors in the assembly process to improve the overall process Project development time is expected to be shortened to 1 month by using the BailAI image inspection platform Since its establishment more than three years ago, Claireye Intelligence has accumulated a considerable amount of project experience and hopes to commercialize the project experience Shirley Liu pointed out that the trial version of BailAI image inspection will be completed this year 2022 Customers can choose industrial cameras or video cameras based on the detail of the object being inspected It can even use X-rays to capture images, and then the images are automatically marked by the platform Claireye Intelligence will provide customers with AI application models suitable for the field Inferences can also be made in the cloud or terminals for launch in the manufacturing industry The metals industry, metal casings of industrial computers, connectors, electronic peripherals, and mechanical parts can all use the platform for defect detection and object identification Claireye Intelligence will continue to improve its technical capabilities, accumulate customer experience to complete commercialization, and also accelerate the implementation of AI inspection applications In the mid-term, it will build terminal and cloud infrastructure and shorten the development time of enterprise AI projects from 6-12 months to 1 month, reducing usage time and lowering the threshold for enterprises The long-term goal is to target the Southeast Asian market where Taiwanese businesses are gathered, expand software and hardware integrated AI solutions to overseas markets, and expand the scale of operations

這是一張圖片。 This is a picture.
AI Smart Health Prevention Plan

Herji Ltd held an interactive teaching session with AI storybooks at the 'Taiwan Early Childhood Development and Remedial Association Taitung Office', allowing children, teachers, and parents to engage in immersive educational experiences AI-generated children's educational storybook materials AI Learning Platform In recent years, changes in the social structure of Taiwan, combined with experiences in hospital emergency departments, have often led us to overlook the depressive symptoms exhibited by adolescents, resulting in tragic incidents of self-harm or even suicide among children A significant part of children's depression stems from their academic performance, with parents worrying about their children's future competitiveness, thus placing a lot of pressure on children who perform poorly academically In a family with two children with the same genetic background and provided with the same resources for growth, we often find that the second child's academic performance is not up to par, with poor grades, an inability to concentrate in class, and even lacking the patience and perseverance to finish reading a comic book or playing a video game We have been exploring why these differences occur and discovered that these issues often arise from undiagnosed learning disabilities during early childhood Due to environmental factors, children with delayed learning abilities are often not acknowledged by over 80 of parents who are reluctant to seek treatment for their children, primarily fearing that their child will be labeled as delayed As a result, the child's learning ability is hindered from an early age, with their academic struggles increasing as they enter primary and secondary school, leading to greater academic lag, frustration from parents, struggles from the child, and increased family disputes If tutoring does not yield effective results, expenditure without achieving positive outcomes often leads to further family conflicts, creating a vicious cycle that accumulates a lot of negative emotions in children during their developmental process, which in turn affects various factors impacting their health In reality, the main reasons behind a child's poor academic performance, inability to learn, lack of interest in learning new things, or even developing health-impacting psychological conditions, actually stem from accumulated learning delays during early childhood The period before the age of six is considered the golden window for treating learning delays If these can be identified and addressed during this time, there is a chance that the child's learning abilities can be greatly improved10The industry's current pain points are as follows 1Lack of methodologies for assessing learning abilitiesLack of databases for sample comparisons in the market 2Traditional parental misconceptionsFear of labeling and treatment delays for mild to moderate cases 3Lack of therapeutic materials and toolsShortage of therapeutic storybooks and series courses This project will develop a national talent development support system, utilizing AI Technological development of a system for assessing children's learning abilities that supports parents in safeguarding their children's health from the start of learning ability testing, offering early detection and treatment In the future, all Taiwanese children, regardless of background, will be able to establish a healthy foundation in early childhood, growing up to become valuable assets for national development 2、 As proposed in this planAIApplications and explanations 'Child Language Ability'AI'Analysis Model' This model quantitatively analyzes 'the condition of children's use of Mandarin' when 'expressing an event' Scenario Preschool teachers guide children in narrating storybook contentAITools analyze the sentences used by children to describe storybook content, applying statistical algorithms for quantitative analysis Analysis indicators include 'sentence type' and 'lexical items' Analysis aspects include correctness of sentence structures, diversity of vocabulary, quantity of vocabulary used, and accuracy of vocabulary usage Application Comparative analysis between an individual child and peers' language abilities can offer more detailed language skills teaching by preschool teachers Techniques used Chinese word segmentation, Chinese POS tagging, Chinese syntactic rules analysis algorithm, and quantitative analysis algorithms Tools usedChinese word segmentation tools, POS tagging toolsChinese POS Tagging Tool 3、 Expected Industrial Value Establish a learning ability assessment and support system, through therapeutic storybooks and courses Collaborate with kindergartens to develop learning ability bases, preventing children from being stuck at the starting point Alongside parents, protect children's health starting with learning ability testing, backed by a robust database, allowing parents to identify early any delays in learning, helping children regain their learning abilities 4、 Expected Industrial Benefits Economic Benefit and Future Spread and Impetus By supporting children with delayed learning abilities, enhancing their learning prowess through this project, these children serve as the future of our nation and can thus significantly contribute to national talent development Furthermore, the purpose of establishing a learning ability development base is to help reunite children with their parents, increasing their interaction time, allowing the children to move beyond mere one-dimensional interactions 3C This facilitates two-way interactions between the child and parents, potentially impacting children who may have been otherwise delayed in developing their capabilities due to environmental factors 「Translated content is generated by ChatGPT and is for reference only Translation date:2024-11-15」