Design/Pseudo Code for Redis-based Ecommerce Recommendation App

  • Status: Closed
  • Prize: $440
  • Entries Received: 1
  • Winner: rifatsikder333

Contest Brief

I am fleshing out an app for making product recommendations for an e-commerce site. I have a rough outline of the architecture but I am looking to define microservices and the Redis services that will implement them. Delivery is not code but pseudo code that shows how Redis will be used in various ways in stages of the app. The app can refer to other non-Redis services such as Spark or anything else.

Recommended Skills

Employer Feedback

“Rifatsikder333 really understood my true needs and helped meet them.”

Profile image evolvedmediavw, United States.

Top entries from this contest

View More Entries

Public Clarification Board

  • evolvedmediavw
    Contest Holder
    • 11 months ago

    Part 3:

    #For each domain we have a machine learning model that will recommend the most suitable items
    #according to the users feature vector. The models are trained on Spark and stored on Redis ML.
    for domain in domains:
    #for each item class of the domain get a score from the random forest classifier
    for item in domain.items:
    domain.itemScores[item] = RedisRandomForestClassify(forestId = "domain:item", featureVector)

    #sort the classification results and get the top results to render recommendations
    results[domain] = domain.itemScores.sort()[0:n_items]

    • 11 months ago
  • evolvedmediavw
    Contest Holder
    • 11 months ago

    Part 2:

    #We should have a domain object which will contains the items and attributes of the domain.
    #The domains will be stored in a dictionary. e.g.:
    #domains[news] = ("World", "U.S.", "Politics", "Business", "Tech", "Science", "Health", "Sports", "Arts", "Style", "Food", "Travel")
    #domains[ads] = ("Nissan", "pertasperlite", "panachemodels", "TransMilenio", "hermaspuspito", "Jeep", "CocaCola", "Konami", "Samsung", "BWWings", "WWENetwork", "givenchy")
    #The domain items will be loaded from a Redis set:

    domains[news] = redis.smembers("domain:news:items")
    domains[ads] = redis.smembers("domain:ads:items")

    • 11 months ago
  • evolvedmediavw
    Contest Holder
    • 11 months ago

    Here is the sort of code I am looking for:

    MS: Real Time Recommendation

    #Load the user's feature vector from Redis. The feature vector is stored as a redis hash.
    #an example of a feature vector: ("age:31", "sex:male", "food_1:pizza", "food_2:sriracha", "last_url:news.ycombinator.com")
    featureVector = redis.hget(uid:features)

    • 11 months ago
  • anmsaiful
    anmsaiful
    • 12 months ago

    I have experience in both design and development. But I am not clear about your project specification. Feel free to initiate a chat to discuss about the project in detailed. Hope I can come up with a working solution. :-)

    • 12 months ago
    1. evolvedmediavw
      Contest Holder
      • 12 months ago

      Did you see the document? What I'm looking for is to write pseudo code that shows how a set of microservices would work, invoking Redis capabiliites, to implement a generic e-commerce recommendation/personalization engine.

      • 12 months ago
    2. evolvedmediavw
      Contest Holder
      • 12 months ago

      The deliverable would be Pseudo code for the microservices defined on the document associated with this project. The idea is to define what Redis services would help implement each microservice.

      • 12 months ago

Show more comments

How to get started with contests

  • Post your contest

    Post Your Contest Quick and easy

  • Get tons of entries

    Get Tons of Entries From around the world

  • Award the best entry

    Award the best entry Download the files - Easy!

Post a Contest Now or Join us Today!