[Data-Sci] [UCI ML] Week 1 Tasks

decks/association_mining.mdx

@@ -0,0 +1,64 @@
+import { CodeSurfer as Surfer } from "code-surfer";
+import { CodeSurferColumns, Step } from "code-surfer";
+import { Appear, Background } from "gatsby-theme-mdx-deck";
+import * as L from "../src/layout";
+import customTheme from "../src/theme";
+import GreetingLoader from "./src/greeting-loader";
+
+export const theme = customTheme;
+
+<L.Column>
+
+# Association Mining
+
+</L.Column>
+
+---
+<L.Column>
+
+# Introduction
+
+* Our photos and introduction about ourselves
+
+</L.Column>
+
+---
+<L.Column>
+
+## Profitable Business
+
+- Businesses need ways to gain profits from the customers by increasing their sales.
+- By selling the product that the customer wants, customer satisfaction and business profits both increase.
+
+</L.Column>
+
+---
+<L.Column>
+
+## Association Mining
+
+- It is a rule based pattern mining technique.
+- It finds features which occur together and are related to each other.
+
+</L.Column>
+
+---
+<L.Column>
+
+## Example: Diaper and Beer 
+
+* When a customer buys diapers, surprisingly, it is highly likely that the customer will also buy beer.
+* This is one of the frequently bought products together and such patterns are found using Association Mining.
+
+</L.Column>
+
+---
+<L.Column>
+
+## Use Cases
+
+* It can be used by stores to arrange their products on racks, i.e. most frequently bought together products besides each other.
+* It can be utilized for carrying out a promotional pricing/sale event.
+* Also used for recommending products to the customers on online platforms.
+
+</L.Column>

misc/proposals/Association_Mining_Proposal.md

@@ -0,0 +1,55 @@
+# Workshop Proposal
+
+### Specific Topic
+_Recommending frequently bought products to potential end-users using Association mining._
+
+
+
+### Interactive Component
+_Each workshop must have an original, interactive component._
+_Calculating the frequently bought items and every correct answer will get a reward._
+_This will help the students to understand the working behind the apriori algorithm._
+
+
+
+### Code Deliverables
+_In Weeks 2-3, your duo must code out the interactive component, as well as any other code needed for the entire workshop. 
+What coding tasks do you need to complete? Each task should also come with what week each task will be completed as well as who is responsible for each task.
+Please be specific._
+_Also, you can use Week 4 to complete code deliverables. Note that this is highly **discouraged** because Week 4 is reserved for finishing touches and preparing the presentation._
+
+- [ Preprocessing data]
+- [ Association Mining using the Apriori Algorithm]
+- [ Trick to reduce the huge number of pairs generated in the algorithm]
+- [ Formatting the output]
+
+### Slide Outline
+_What is a general roadmap for your slides?_
+
+* Define the dataset and what features we choose
+* About Association mining
+* How will we use the algorithm to detect the frequently bought items
+* Other algorithms which students can research
+
+### Timing
+_Each workshop should be around an hour to 90 minutes. How long will your workshop be?
+_Planned to be precisely one hour on the dot but might run up to 90 minutes depending on the Q and A section._
+_Preprocessing the data will take less amount of time approcimately 10 minutes as the data is already clean._
+_The algorithm will take 40 minutes to explain but the running time might vary depending on machines._
+_Last 10 minutes will be allocated to discussing other algorithms and QnA section._
+
+
+### Motivation behind project
+_What motivated your duo to pursue this idea?_
+_Businesses need to expand their business and one of the crucial components of any successful business is customers and their needs. Recommending the customers what are the frequently bought goods together is a good way to enhance the business sales._
+
+
+### Explain difficulty/prior experience needed
+_Is there any prior experience needed for this workshop? Please be specific._
+_Python (Pandas, Numpy libraries)_
+
+
+### Technologies that will be used
+_Please be specific.
+* Python
+* PySpark