Lesson 1Planning Transaction Tables: Orders, Order Items, Returns, Lifetime Value Signs and Field ChoicesLearn to plan main tables for transactions that record orders, items in orders, returns, and signs of lifetime value for customers. We talk about important fields, ways to organize data without repeats, and how to help analysis later and work for recommendations.
Order header vs line item schema designModeling returns, refunds, and cancellationsCapturing discounts, coupons, and taxesStoring lifetime value and margin signalsKeys, indexes, and partitioning choicesLesson 2Dealing with Noisy and Thin Behavior Data: Making Sessions, Filtering Bots, Removing Duplicates, Weighing EventsLook at ways to clean logs of behavior that have noise and make thin data useful. You will learn rules for making sessions, filtering bots and scrapers, logic to remove duplicates, and ways to weigh events made for training recommendations.
Sessionization rules and timeoutsDetecting and filtering bots and scrapersClick, view, and purchase deduplicationEvent weighting for model trainingHandling sparse users and cold startsLesson 3Planning Product List Table: Product ID, Title, Category Tree, Features, Price, Brand, Stock, Pictures, Standard Text, EmbeddingsLearn to build a product list table that helps quick finding and rich recommendations. We cover IDs, features, prices, stock levels, media files, standard text, and embeddings, plus ways to update and avoid too much organization for speed.
Stable product and variant identifiersCategory hierarchy and attributesPrice, stock, and availability fieldsImages, media, and canonical textStoring and updating item embeddingsLesson 4Rules for Making Features for Recommendations: Newness, How Often, Money Spent, Item Popularity, Category Likes, User EmbeddingsFind main rules for making features in systems that recommend. We explain newness, how often things happen, money value, popularity of items, likes for categories, and user embeddings, including time windows for grouping and safe ways to compute without mistakes.
Recency, frequency, and monetary featuresItem and category popularity signalsUser–category and brand affinity scoresSequence‑based and session featuresUser and item embedding generationLesson 5Extra Data Sets: Item Details, Category System, Promotions, Content (Descriptions), Supplier InformationUnderstand how extra data sets make recommendations better than just clicks and orders. We cover details on items, category systems, promotions, content like descriptions, and info from suppliers, plus how to keep them matching, versioned, and easy to combine when big.
Designing item metadata schemasMaintaining product taxonomy hierarchiesModeling promotions and price rulesStoring rich content and descriptionsIntegrating supplier and feed dataLesson 6Ways to Clean Data and Fill Gaps: Missing Features, Strange Prices, Wrong Time StampsLearn real ways to clean data and fill in missing parts for e-commerce. We deal with missing features, odd prices, wrong time stamps, and currencies that do not match, focusing on rules, simple tricks, and how they affect quality of recommendations.
Detecting and fixing missing attributesHandling outlier and zero pricesCorrecting invalid or noisy timestampsCurrency, tax, and unit normalizationDocumenting cleaning rules and impactsLesson 7Planning Event Flow and Interaction Table: Event ID, User ID/Session ID, Event Type, Product ID, Time Stamp, Context (Referrer, Page Type), Device, Event ValuePlan a single table for interactions and flow of events that catches user actions across different ways. Learn structures for events, IDs, context fields, and how to help both live streaming and batch work for recommendation lines that are not online.
Choosing event and user identifiersModeling event types and propertiesCapturing context, device, and referrerEvent time, ingestion time, and orderingStreaming vs batch storage patternsLesson 8Planning User Profiles Table: Key Fields (User ID, Signup Time, Email Hash, People Details, Life Stage, Groups, Opt-In Signs) and ReasonsPlan a table for user profiles that balances power for personal touch with keeping privacy and following rules. We cover key fields, life stages and groups, opt-in signs, hashing sensitive info, and how profiles help models for recommendations.
Core identifiers and signup metadataDemographics and lifecycle stagesBehavioral and marketing segmentsConsent, opt‑in, and preference flagsPrivacy, hashing, and retention rules
