[{"child":[],"type":"css","text":"@import '/static/less/ncp.less';\n@import '/static/less/asset.less';\n\ndiv.full {\n width: 100%!important;\n margin:0!important;\n &.bg {\n background-color: @c-bg;\n }\n}\n","value":"/* font */\n/* colors */\n/*\n0 : 모바일 상품 상세 상단에 < 네모 컬러\n1 : 상품목록아이템 상단 바\n2 : 스크롤메뉴 비활성화 컬러\n*/\n/* button */\n/*--------------------------------------------------------------------------------------------------------------\n * Style\n *--------------------------------------------------------------------------------------------------------------*/\n.flex-vhm {\n display: flex;\n justify-content: space-between;\n align-items: center;\n}\n.flex-hm {\n display: flex;\n justify-content: space-between;\n}\n.flex-vm {\n display: flex;\n align-items: center;\n}\n/*--------------------------------------------------------------------------------------------------------------\n * Template\n *--------------------------------------------------------------------------------------------------------------*/\ndiv.full {\n width: 100%!important;\n margin: 0!important;\n}\ndiv.full.bg {\n background-color: #f6f7f8;\n}\n","uId":"16302973174061","active":false},{"child":[],"type":"tab","idName":"","className":"col-4","data":[{"link":"","text":"Benefits","blank":false},{"link":"function","title":"","blank":false,"text":"Features"},{"link":"price","title":"","blank":false,"text":"Pricing"},{"link":"${guideUrl}/en/aitems-overview","title":"","blank":true,"text":"Documentation "}],"uId":"15699065211272","active":false},{"child":[{"child":[{"child":[],"type":"title","idName":"","className":"","text":"### Easily build sophisticated personalization capabilities\nAI trend in which you can easily create and test personalized recommendations even without machine learning expertise.","uId":"15687730540678","active":false,"useYn":"Y"}],"type":"section","idName":"","className":"wrapper","uId":"15687730456665","active":false,"useYn":"Y"},{"child":[{"child":[],"type":"text","idName":"","className":"","text":"Personalized item recommendations\n: AiTEMS lets you recommend the items that are most likely to attract users' attention by matching meta-information of products or information objects against meta-information of users stored in profile. Create recommendations even for new users and products with insufficient data. You can respond to changes in customers’ clicks and purchase intention.","uId":"15687731992269","active":false,"useYn":"Y"},{"child":[],"type":"text","idName":"","className":"","text":"Reduce development and operation costs\n: You can benefit hugely in terms of time and costs if you choose to use AiTEMS over building your own system. AiTEMS allows you to easily create and test personalized recommendations even without machine learning expertise. You pay for what you use only when you need the service. ","uId":"163063922495510","useYn":"Y","active":false},{"child":[],"type":"text","idName":"","className":"","text":"Personalized ranking from user preference\n: AiTEMS analyzes user data such as user activity logs and preferences and optimizes personalization models. Based on a set of items learned from user's data sets, AiTEMS creates a personalized ranking and recommend products tailored to the user's taste or needs.\n","uId":"163063922585111","useYn":"Y","active":false}],"type":"section","idName":"","className":"function","uId":"15687730479636","active":false,"useYn":"Y"}],"type":"div","idName":"","className":"full bg","text":"","data":{"sm":"","md":"","lg":"","xl":""},"useYn":"Y","active":false,"uId":"16302973279232"},{"child":[{"child":[],"type":"title","idName":"","className":"","text":"### Features\n","useYn":"Y","active":false,"uId":"16798930702249"},{"child":[],"type":"text","text":"### Fully managed machine learning service\nThe service contains all aspects of the recommendation technology necessary for machine learning, ranging from data preprocessing and training to calibration and prediction.\nBy loading user records and clicking a button, you can generate recommendations based on data trained and analyzed by AI machines.","uId":"16798913857701","active":false,"useYn":"Y"},{"child":[],"type":"image","idName":"","className":"mt-30","alt":"","data":{"sm":"","md":"","lg":"","xl":"https://xv-ncloud.pstatic.net/images/product/img-en-ai-tems-01@2x_1639979756593.png"},"useYn":"Y","uId":"16306396739915","expandYn":"N","active":false},{"child":[],"type":"text","text":"#### Strong data security\nBoth the service user's data uploaded by the service provider for data training and the outcomes produced by the AI system's learning process are owned by the service provider. All data you upload is kept private and encrypted, and used only when creating recommendations for your customers.\n","useYn":"Y","uId":"16306396739914","active":false},{"child":[],"type":"text","text":"#### Recommendation model types\nAI machine offers three recommendation model types including \"Recommended for you\", \"Others you may like\", and \"Most purchased\".

\n\\- Most purchased: recommends the items most purchased by rank
\n\\- Recommended for you: predicts the product that a user is likely to purchase, based on past purchases, views, and requests of that user
\n\\- Others you may like: predicts the product that a user is likely to prefer by identifying their true intentions based on a user’s past clicks and purchases\n","useYn":"Y","uId":"16798924994833","active":false},{"child":[],"type":"image","idName":"","className":"mt-30","alt":"","data":{"sm":"","md":"","lg":"","xl":"https://xv-ncloud.pstatic.net/images/product/img-en-ai-tems-02-2-nd@2x_1639979766038.png"},"useYn":"Y","uId":"16313691723918","expandYn":"N","active":false},{"child":[],"type":"text","text":"#### Hyperparameter optimization\nYou can make use of Hyperparameter Optimization (HPO) to improve the performance of your machine learning model. \nThe HTO settings paly an important role in the prediction accuracy of a machine learning model.","useYn":"Y","active":false,"uId":"16798928512036"}],"type":"section","idName":"","className":"wrapper function","useYn":"Y","active":false,"uId":"163029800109824","expandYn":"N"},{"child":[{"child":[],"type":"title","idName":"","className":"","text":"### Pricing information\nThe usage fees for AiTEMS are charged by adding the training time and the costs for recommendation results.
\n※ Training time may take up to 24 hours depending on the recommended type.
\n(e.g., 24 hours each for up to 48 hours when applying for two recommended types)","uId":"16391170292112","useYn":"Y","active":false},{"child":[],"type":"priceDataTableFilter","options":{"type":["region","currency"]},"useYn":"Y","active":false,"uId":"16588910813145"},{"child":[],"type":"priceDataTable","vat":"Y","search":"N","data":{"productCode":["AITEM.STAND.V001","\nAITEM.TRIAL.V001"],"productRatingTypeCode":[],"noUniq":true,"rowItems":[{"headerText":"Service Plan","width":"200px","tag":"th","itemKey":"productCode","unit":"","orderKey":"","codeNameList":["AITEM.STAND.V001: Standard","AITEM.TRIAL.V001: Trial"],"orderBy":"","orderNo":""},{"headerText":"Type","width":"200px","tag":"th","itemKey":"productRatingTypeCode","codeNameList":["RCMD_RSLT_BATCH: Recommendation","EDUCATION_TIME: Training","RCMD_RSLT_REALTIME: Real-time recommendation"],"unit":"","orderKey":"","orderBy":"","orderNo":""},{"headerText":"Billing section","width":"auto","tag":"td","itemKey":"rangeText","unit":"Count","orderKey":"","codeNameList":[],"orderBy":"","orderNo":""},{"headerText":"Billing standard (Month)","width":"auto","tag":"td","itemKey":"priceUnitCodeName","unit":"","orderKey":"","codeNameList":[],"orderBy":"","orderNo":""},{"headerText":"Usage fee","width":"auto","tag":"td","itemKey":"priceText","unit":"","orderKey":"","codeNameList":[],"orderBy":"","orderNo":""}]},"useYn":"Y","active":false,"uId":"16588910893026","maxHeight":"Y"},{"child":[],"type":"custom","text":"

Training time is rounded up by 1 hour and billed per hour. (For example, if 50 minutes are used, it will be rounded up to 1 hour.)

Data storage space for learning uses Object Storage, and its billing follows the Object Storage pricing policy.

The recommendation results of the Standard plan are based on the learning of a general Batch method, and the model should be updated by requesting relearning each time to reflect the changed data after the first learning.

The real-time recommendation results of the Standard plan are based on the learning in which a general Batch method and an incremental learning feature are added, and after the first learning, the results are automatically updated to the model according to the specified interval.

The real-time recommendation results of the Standard plan are optional and Batch method learning is operated once a day when the feature is selected. It is charged by adding the number of recommendation results for items updated in real-time.

The real-time recommendation results of the Standard plan can be suspended at any time, and the learning will be suspended after a suspension is requested.

For more information on the AiTEMS service, refer to the introduction for the AiTEMS service.

","useYn":"Y","uId":"16391170292124","active":false},{"child":[],"type":"title","idName":"","className":"","text":"

Pricing examples

\n Case 1) The company spends 50 hours to complete the training of Batch pricing plan and provides recommendation results for 1 million users.\n

\n - Training fee: 50 training hours x 2500 KRW per training hour = 125,000 KRW
\n - Recommendation result fee: 1,000,000 results x 70 KRW/1,000 recommendation results = 70,000 KRW
\n - Total cost: 125,000 KRW + 70,000 KRW = 195,000 KRW

\n Case 2) The company spends 5 hours per training through the \"Standard pricing plan\", performs a single learning a day, and provides an average of 270,000 recommendation results when performing a single learning.\n

\n - Training fee: 5 hours x 30 days x 2500 KRW per training hour = 375,000 KRW
\n - Recommendation result fee: 270,000 results x 30 days x 70 KRW/1,000 recommendation results = 567,000 KRW
\n - Total cost: 375,000 KRW + 567,000 KRW = 942,000 KRW

\n Case 3) The company selects the \"real-time recommendation option\" for the \"Standard pricing plan\", spends 5 hours per training, performs a single learning a day, and provides an average of 270,000 recommendation results when performing a single learning.
And the company provides the daily accumulated average of 1,000 update results as the real-time recommendation results.\n

\n - Training fee: 5 hours x 30 days x 2,500 KRW per training hour = 375,000 KRW
\n - Recommendation result fee: 270,000 results x 30 days x 70 KRW/1,000 recommendation results = 567,000 KRW
\n - Real-time recommendation update fee: 1,000 results x 30 days x 200 KRW/1000 recommendation results = 6000 KRW
\n - Total cost: 375,000 KRW + 567,000 KRW + 6,000 KRW = 948,000 KRW

\n","useYn":"Y","active":false,"uId":"16689960047981","expandYn":"N"},{"child":[],"type":"title","idName":"","className":"","text":"

Pricing sample

The usage fees for AiTEMS are charged by adding the training time and the costs for the number of recommendation result execution by calculating them in units of 1000 items according to the segment.\n

- Usage fees = (training time * hourly usage fee) + (recommendation results/1000 * usage fee per 1000 items)\n

If the accumulated monthly training time is 70 minutes and the recommendation results are 100,000 items, then the monthly fees are as follows.\n

- (2h * {{ _price([{ c:'AITEM.STAND.V001', r: 'EDUCATION_TIME', a: 1 }]) }}) + (100,000/1,000*{{ _price2([{ c:'AITEM.STAND.V001', r: 'RCMD_RSLT_BATCH', a: 1000, p: 0 }]) }}) = {{ _price2([{ c:'AITEM.STAND.V001', r: 'RCMD_RSLT_BATCH', a: 100000, p: 0 },{ c:'AITEM.STAND.V001', r: 'EDUCATION_TIME', a: 2 }]) }}\n\n

\n\n","useYn":"N","active":false,"uId":"16391170292125"}],"type":"section","idName":"","className":"price","uId":"16391170292111","active":false,"expandYn":"N"}]

AI

Hybrid

Finance

Game

Global Service

Live commerce

SMARTWORK

AiTEMS