CLOUD SERVICE APPLICATION CASES

AI Cloud Service

The linkage between KT GIGA Genie and ADAMS.ai

AI Speaker – a means to help people experience AI technologies easily in everyday life, has been introduced throughout the world. In 2014, Amazon launched “Echo” – a product based on the AI virtual assistant Alexa, which is considered the first product ever that helps the related market grow rapidly.

In Korea, three mobile carriers KT, SK Telecom, and LG Uplus, are leading the AI speaker market. This market is growing rapidly despite being a latecomer compared to major countries’ markets led by global Internet companies such as North America. Among these mobile carriers, KT is the No. 1 company in the domestic AI speaker market with 1.5 million subscribers. Since the introduction of the AI speaker “GIGA Genie” in January 2017, KT has been diversifying its product lines into GIGA Genie LTE, GIGA Genie Buddy, GIGA Genie 2, and AI Makers Kit.

GIGA Genie operates closely with the KT home service platform (IPTV, music, phone call, home IoT). It is linked with platforms by third parties, composing an overall AI ecosystem. Among the third parties’ platforms, GIGA Genie links with Saltlux’s ADAMS.ai platform to advance dialog quality between a user and AI. This provides a lifelike Deep Q &A service that requires extensive knowledge within such a linkage.

① Main Content

GIGA Genie provides a Q&A service that finds and delivers answers to users’ questions. GIGA Genie is linked with Saltlux’s ADAMS.ai for Deep Q&A in fields that include real-life information, general knowledge, and expert with extensive background knowledge. When a user’s question submitted through GIGA Genie is delivered to the ADAMS.ai platform, ADAMS.ai finds an answer to the question through large-scale knowledge graph-based Deep Q&A technology. The answer is then delivered to GIGA Genie and provided to the user. In this process, new knowledge vital to Q&A is added and learned continuously to improve the quality of the service.

② Applied technology

1. Flexible Q&A using the ensemble technology
   KT’s GIGA Genie finds answers to equations in fields that include real-life information, general knowledge, and expert knowledge using the Deep Q/A service provided by ADAMS.ai through the Open API. The Deep Q/A service provides answers to submitted questions by forming an ensemble of knowledge graph-based Q&A (KBQA), information search-based Q&A (IRQA), MRC-based Q&A (MRQA). It then selects the optimum solution according to the nature of each particular query and answers the user.
2. Continuous Q &A monitoring and knowledge curation
   Continuous Q&A monitoring across fields that include real-life information, general knowledge, and expert knowledge where Deep Q&A is applied. Adding and establishing large-scale new knowledge in each field every day through knowledge curation. The established new knowledge is converted into data to use for Q&A through the management function of the platform.

③ Main performance

It provides a complex reasoning speed of 500,000 knowledge units per second, utilizing general knowledge of 800 million items established as a knowledge base. It also delivers a 94% probability of choosing the correct answer in real-life information, general knowledge, and expert knowledge through complex reasoning and knowledge learning from 5 million documents – the quantity of data collected each day.

The linkage between Woori Bank AI consultation system and ADAMS.ai

Recently, various customer service channels, including mobiles, have been expanded in financial fields. In addition, there is a growing demand for a service responding to the latest AI-based technology trends. Woori Bank also saw the need to establish an AI consultation system for customers and employees in response to a paradigm shift in financial services.

In September 2017, Woori Bank initiated the ChatBot service ‘WebeeBot’ that enables real-time consultation using AI technologies. ‘WebeeBot’ provides financial consultation services 24 hours a day. It provides an answer by capturing the intent of the questioner and chatting with the customer just like that of a human consultant, instead of using the question selection method and pre-order method. It also provides general information besides financial information.

① Main Content

Woori Bank’s AI Consultation System can be divided into the data area, AI platform area, and service area. It is linked with Saltlux’s ADAMS.ai, a huge knowledge base-based platform, to provide general information through Q&A, such as weather and personal information. Knowledge of consultation services in financial fields is internally established in the data area and used in consultation services. In the case of general information that cannot be obtained within a short time, it is serviced through a link with ADAMS.ai – an external open platform.

② Applied technology

1. Knowledge graph-based Deep Q &A
   Woori bank’s ‘WebeeBot’ finds answers to general knowledge questions as well as financial-related questions using the Deep Q&A service provided by ADAMS.ai through the Open API. The Deep Q&A service provides Q&A about current general knowledge through complex reasoning based on Asia’s largest knowledge graph.
2. Prompt application of new knowledge
   Providing a platform for managing information, such as dictionaries, knowledge, and indexes. It enables Q&A processing by creating, learning, and reasoning new general knowledge. Deep Q&A is applied so that new knowledge can be integrated and corresponding Q&A can be implemented.

③ Main performance

Woori Bank’s ‘WebeeBot’ is linked with ADAMS.ai, which embeds knowledge graphs for approximately 800 million general knowledge items. It not only provides consultation in the financial field, but it also extends to general information areas, such as weather and personal information.

Data Science Cloud Service

Status of Data Science Cloud Service

Korean Women’s Development Institute – Collection and analysis of social big data regarding policies for women

Due to the development of data-related technologies, the research form is shifting from sampling surveys – traditional investigation methods of the past, to big data-based complete surveys. This spreads to all industrial fields, including public institutions, local governments, and companies. Various policies for providing big data-based administration services and establishing evidence-based policies are being promoted. Therefore, the intention is to generate basic data to determine the demands for policies for women and prepare countermeasures by analyzing public opinion in a society that changes minute by minute.

① Main Content

This research has been promoted to analyze various gender-based social phenomena and issues on social media by collecting and analyzing social data. The analysis method uses word value analysis, emotion analysis, and word analyses, which were internal solutions from Saltlux. The analysis result using the visualized chart was evaluated as an important research result since it provided basic data for promoting policies for women.

② Applied Technology

1. Trend Analysis
   Determine the main interests of people. Whether the discussion was initiated, expanded, or reduced was based on the frequency of reference concentrated by aggregating the frequency of analysis target references.
2. Word weighted value and emotion analysis
   Words that appear frequently in documents are filtered. Important words are extracted using the frequency of word appearance within a document and across documents. Emotions within each sentence are classified using the emotional word dictionary and map machine learning. Based on the findings, users can determine the comprehensive emotions of the analysis target or word.
3. Concurrently appeared in word and original text (VOC) analysis
   Important words that appear frequently are sorted and only sentences with relevant words are extracted. Original text, where keywords are mentioned together, is extracted to determine the correlation between those keywords. Trends and linkage contexts of an analysis target are determined by examining extracted original text.

③ Main Performance

In this research, data from press media, social media, and communities related to subjects (feminism, minimum wage) are collected in a comprehensive public opinion analysis with a limited understanding of a certain sample. Also, basic data to promote policies for women are prepared by analyzing the collected data. This is a research case that extends research in media that used to be scarce and is expected to promote the future use of big data by discovering the demands for policies for women and seeking countermeasures.

Hyundai Motor Company – Worldwide competing vehicles information collection

The worldwide competing vehicle information collection service can improve customer satisfaction and reinforce market competitiveness. It is done by preparing external data in multiple languages from multiple channels and active market sensing. For that reason, an exclusive data collection infrastructure is constructed to understand customers’ needs, acquire trending information, monitor, and extract potential risks from the collected information in real-time. Data are collected from news, magazine, cafe, community, vehicle-related forum websites, social media, and websites requested by customers. The collected data is standardized (automatic classification, filtering, purification, processing), established, and provided in a form that customers can use immediately.

① Main Content

This project was carried out for the purpose of improving customer satisfaction and enhancing market competitiveness through the collection of external data using multiple channels and active market sensing using the collected data. For such purposes, an exclusive data collection infrastructure that enabled a comprehensive understanding of customers’ needs, acquisition of information regarding the trends of the competitive market, real-time monitoring, and extraction of potential risks from the collected information was constructed. The target data for collection include data from vehicle-related detailed message boards in websites, cafe websites, keyword-based data and data from Facebook, Twitter, Instagram, YouTube and other custom websites, and the collected data are filtered, standardized and provided in a form that could be used immediately by the customer.

② Applied Technology

1. Data collection
   Approximately 1,500 target websites for collecting selected by the client are examined and sorted. The curation center performs metadata investigation for the target websites that collect and filter data through consultation with the client. After extracting search result URLs and original text URLs, the frequency is analyzed. The curator also finds a website directly related to the vehicle.
2. Data quality automation
   Data filtering, automatic cafe schema detection, machine learning spam classification, data function standardization, creation of data to transfer, and failure detection are carried out.
3. Collection status monitoring dashboard
   Real-time monitoring of collection status in the project, each project management status verification, top website verification within the collection risk degree, collected data trend verification, and data collection places verification are available.

③ Main performance

Through the collected data, it provides personalized data that can be used immediately for internal analysis through a comprehensive understanding of the client’s needs, acquires trending information of the competitive market, monitors potential risks, and extracts hidden information, such as the client’s behavior patterns.

AIA Group Limited – Determined the needs of singles aged between 35 and 45 regarding insurance products and planned insurance products targeting these singles

It was necessary to improve market competitiveness by planning a product based on data rather than the previous method, which depended on previous intuitive experience. It was intended to develop an insurance product that could attract the attention of singles aged between 35 and 45. This could be done by securing targeted analysis data related to single individuals and carrying out convergence analysis and cognitive analysis of structured and unstructured data.

① Main Content

This project was an example of big data collection and analysis by age bracket and gender with the aim of “determining the needs of singles aged between 35 and 45 regarding insurance products and developing an insurance product for them.” This was a case where insight was used to establish a product strategy by AIA Group Limited for singles aged between 35 and 45, and verify issues and hypotheses.

② Application Function

1. Hypothesis verification scenario
   Consumers’ interests and reputations are analyzed by identifying subject keywords and grouping identified keywords to develop an insurance product that can attract singles aged between 35 and 45.
2. Analysis of reference frequency trends
   Determine trends and tendencies of customers’ interests, and examine detailed issues through trend comparisons for each property.
3. Word Cloud and Keyword Analysis
   Provide insights through consumers’ interests and reputation analysis for each related issue, and identify the relationships between related keywords.
4. Sentiment analysis
   Determine consumers’ responses compared to the appearance frequency through emotion analysis results of detailed property keywords for each property and a bubble chart analysis.

③ Main Performance

It helps improve customer satisfaction and market competitiveness through active market sensing using external data from multiple channels. This verified platform enables users to develop a customized product just by changing the target.

PwC Consulting – Real-time company information and monitoring service

It is essential that a company promptly responds to rapid social change by determining the company status in real-time. Especially, collecting, analyzing, and visualizing huge social media and SNS data are necessary to gain insight into a specific company.

① Main Content

Provide real-time company information and competitor monitoring service based on social data of 15+ billion cases. The analysis functions, including trend analysis, keyword comparison and analysis, related keyword analysis, emotion analysis, and dictionary management are also available, thereby offering an important result in the company analysis and monitoring.

② Application Function

1. Social Data Search
   Extract and provide a document containing a query from each search source of news, blogs, and Twitter. Provide the total number of searched documents and the total number after removing redundant ones.
2. Trend Analysis
   Provide the statistics of documents registered in news, blogs, and Twitter regarding a given query by period. Provide the results in the form of JSON that correspond to the query, and the results of entire document statistics within the same period.
3. Automatic related subject analysis
   Related subjects are extracted from news, blogs, and Twitter data regarding a given query. Provide a ranking based on the appearance frequency and the connection information. Also, the unit duration can be set for a selected duration.
4. Today’s Topic Analysis
   Extract N number of today’s topics for an entered news category and provide its ranking.

③ Main Performance

The entire process, including collecting, storing, analyzing, and expressing data regarding a certain company and competitors can be processed integrally. In this project, press media, social media, and community data related to companies were collected. Various information, including the status, technologies, and service forms of a company is collected. Through the collected data analysis, Insight is provided and used as customized information.

Korea Institute of Science and Technology (KIST) – Data-based R &D environment establishment service

When the percentage of produced digital data increases gradually, the introduction of a research support system using big data from a research institute can improve the institute’s overall productivity. In this product, a global-scale research data analysis environment that could improve KIST’s research productivity was built by establishing a structured and unstructured data collection and management system to maintain competitiveness in R&D.

① Main Content

Establish a system to share internal research information in real-time and link it with external (overseas) research information that sets research direction to assist the research institute’s R&D productivity. This system includes KiRI DataBank which can collect, purify, store, and analyze materials (Korean and foreign papers/patents) for catalyst field-related research. The KiRI Note platform to record and manage research data is also available.

② Applied Technology

1. Inquiry for the recording and status of research activities
   The research note function allows users to efficiently manage research activities and the status of ongoing projects.
2. Basic paper search and analysis, similar papers, trends, and keyword analysis
   This function searches for and analyzes content related to main keywords. Various analyses, including basic time series analysis, trend comparison, keyword analysis, and paper-author or material-paper network diagrams are available.
3. Curation tool that can extract paper information in detail
   Both the basic information of a paper crawled from the web and the curation tool are provided. DB is automatically created using basic abstract information. Detailed and technical contents are structured, stored, and managed through the curation process.
4. Digitizing tools that extract paper analysis results
   The digitizing tool can convert the research results in the form of a graph into data. When selecting an image to extract, users can set the axis and target. Also, it allows converting a graph image into digital data while the result is stored and managed in the storage.

③ Main Performance

Support research through research information sharing and qualitative improvement of research productivity and communication. To improve the latest information sensing and paper for the catalyst field, it establishes DB using Korean and overseas research papers and patents. The experiment notes created through various experiments are converted into data and applied to data science, such as data mining and AI.