Weekly Question: Week 13
Answer one of the following questions as a comment on this post. You must be logged in to comment.
Submit your contribution by Wednesday, November 30. Remember that if the question is not answered by midnight on November 30 you will not receive participation credit for your answer.
- Name and describe a business question that you could answer using associate rules (market basket analysis). What data would you collect to perform the analysis? Don’t use one we’ve covered in class.
- Your Uncle Bob’s store sells food, magazines, cleaning supplies, and toys. Bob knows nothing about technology, but he knows you do! He’s asked you for help, but you want to keep it simple for him. Choose the technique that you think will most help him (decision trees, clustering, or association rules) and explain to your Uncle Bob how it can help his business.
David Schuff
Uncle Bob should use the association rules. With the association rules, he can find out which products he wants to put near each other in the store. For example, we will be able to see how often people buy magazines and cleaning supplies together, and if they have a high lift, it means that they are often purchased together. If the lift is below 1, they are probably bought together by chance.
The most helpful technique for Uncle Bob would be with association rules. It will help him discover the interesting relationships between the products that he sells. The market basket analysis answers questions such as “What items are bought in association with another?” Calculating the strength of the association (confidence) is also relatively easier. The scale is only from 0 to 1, 1 being complete association and 0 being no association. Not only will the association rules keep it simple for Uncle Bob, it will also help him predict the occurrence of the items that he carries. His prediction will help him in many business aspects: he will know which items are bought together, which items are more popular than others, and which items he must carry more in inventory.
Since association rules generally tell us what product combinations are bought and in what order, this could be useful in a large department store. There are a variety of products to be sold there, but placing the shoes near the jeans, or the make-up near the handbags may prove to increase sales if placed correctly.
I agree with this post. Association rules would be the best option because he would be able to track the relationships between his products, arrange them accordingly, and have a better understanding of his customers.
It would be best for Uncle Bob to use association rules. By using association rules, Uncle bob would be able to see the relationship between all of his products. Ideally, it would be beneficial if the market basket analysis was used so we could see what products are bought together and then re-arrange the store accordingly. We could tell which products are associated by calculating the confidence level. An accurate measurement would consist of a confidence level between 0 and 1, with 0 being no association and 1 being complete association. The closer the confidence level is to 1, the stronger the relationship.
Your Uncle Bob’s store sells food, magazines, cleaning supplies, and toys. Bob knows nothing about technology, but he knows you do! He’s asked you for help, but you want to keep it simple for him. Choose the technique that you think will most help him (decision trees, clustering, or association rules) and explain to your Uncle Bob how it can help his business.
Hi Uncle Bob,
I have looked through your previous sales records and put them into computer to generate a business intelligence solution or forecasting for you. From the report, I would say that there are significant association between sales of cleaning supplies and magazines. I also noticed that you intended to put magazines across to your cleaning supplies already. My software told me from its association rules that people usually buy these two together, especially female customers. Association rules are a set of algorithm to calculate correlations between events. It analyzes every sales combination and finds the strong relations in theory for us to test in practice. You could start from this list with the highest number for “lift”, which is an indicator for association. On the other hand, the past data could be predictors for the future, however, it is not always true in real life. Let me know if you have more questions.
In my opinion, the most simple technique to help my Uncle Bob would be showing him the association rules. It would help him strategically organize his store in order to help increase profits. He can see the relationship between each product and rearrange the store to make it convenient for shoppers. This rule is based off of a confidence interval scaled from 0-1. 0 being no confidence between products and 1 being the maximum confidence interval between products. This will show the amount of confidence that certain products will be sold together in the same order. This is easy to calculate and understand so that is why i would reccomend this to my uncle bob.
Uncle Bob’s best choice is association rules. It will keep things simple for him, while still helping him track how much he sells of each item and how they are associated on a 0 to 1 basis with other items. Since it calculates how frequent every possible combinations of items are bought, he will be able to track which items are often bought together which can help with store layout, inventory management, etc.
Uncle Bob should use association rules to help him in his business. Since Uncle Bob sells various types of products in his store, association rules is a great technique to use to analyze relationships between the products sold in his store. In a sense, Uncle Bob can figure out which products predict the occurrence of other items also known as market basket analysis. This analysis can potentially increase sales for Uncle Bob as he will have an idea of what combination of products can bring him the most revenue. Association rules will also help Uncle Bob decide the layout for his store so he can maximize sales from the positioning of the items and their relationships with other products in the store. If Uncle Bob has doubts about the strength of the relationships, associations rules can be a great aid for analyzing confidence otherwise known as the strength of the relationship. A confident relationship will make for a confident association rule. Uncle Bob can focus on various marketing strategies for these strongly related products in an effort to attract more customers while helping the business grow.
The best technique for Uncle Bob would be the technique with association rules. This technique will allow him to discover the relationships between products that he sells his store in the easiest of ways. This technique is the easiest because in order to calculate the strength of the association, use a scale from 0 to 1. (1 being complete association, 0 being no association) The association rules will tell Uncle Bob what product combinations are bought together and in what order the products should be placed.
Association rules is applied for every sales businesses because it measures the proportion of relationship between the products and customers behavior. One of the examples of the market basket analysis that apply in the association rules is the e-commerce business.
For example, customers go to the website and they visit page A and page B, the percentage of confidence predicts that they most likely visit page C. When customers do online shopping for clothes such as dresses (page A), they normally stop by the accessories section to shop accessories that go beyond with the dress (page B). The relationship between page A and page B shows the high percentage of confidence and most likely customers will have a chance to buy shoes (page C)
I would tell Uncle Bob to use association rules to determine the placement of products in his store because his store sells such a variety of products. The association rules give Uncle Bob an idea of the relationships between the products being bought. He could use a market basket analysis to figure out which products to place on promotion and where to strategically place the products in accordance to what products are bought together in the same visit.
The technique I would use to help Uncle Bob for his store would be association rules. Allowing SAS to store and run the data, then graph it, makes it easy to understand what decisions should be made when re-organizing the store to make it more profitable. In using SAS, it allows our stored data to interact with a node, which results in useful information. We would use the graphs in the results to aid us in deciding which items should be placed next to each other in the store for those products to sell more. From the graphs I would look at the highest lift and the rule pair that was associated to that lift. I would start with those products then work my way down the rule descriptions in implementing the changes.
Market basket analysis can apply to a retail stores, one example is Best Buy. Best Buy sells electronics and appliances, and a market basket analysis is a great technique that will allow them to analyze the chances of a service plan or warranty being purchased along with a particular product. They can analyze the percentage of how many customers purchase laptops or televisions with or without a service plan. Also, Best Buy would be able to analyze if warranty plans are purchased the most with appliances opposed to computers.
To answer for question 1,
Amazon online store can be an example to apply market basket analysis. The question to apply the analysis would be how much people get influenced by these following advertisements suggested by Amazon:
*Customers who viewed this item also viewed
*Customers who bought this item also bought
In other words, by seeing whether consumers put suggested products in their shopping carts after the first items they put in the carts, the degree of influence by the advertisements can be analyzed.
The set of data should be collected to perform this analysis is the followings:
*How much % of customers put suggested products after they put their first product in their carts.
*How much % of customers really purchase these suggested products in the order of advertisements.
Bob should use association rules. Association rules are used to predict what items will be purchased together. Using this Bob can find how likely it is that customers will buy 2 or more items together, for example finding how likely it is that people will buy toys with food. Using this information, he can decide where to place items throughout the store. To go back to the previous example, if food and toys are usually bought together, he might want to put them next to each other in the store.
Uncle Bob should use the technique of association rules as a way to improve his business. This is a good choice especially because the products in Uncle Bob’s store are of such a wide variety ranging from things like cleaning supplies to toys. The association rules would help him learn which products best “go together”, which products are likely to be purchased simultaneously. After learning this, Uncle Bob could more easily develop a logical layout for the store, rearranging it according to the relationships described by the association rules. Since Uncle Bob is not to up to date with technology, I believe this is the best system for him considering it is very simple to learn and follow.
Association rules would be the most helpful to Uncle Ben because it could help him sell more products. Association rules show the relationships between different variables. If the data shows that every customer who comes into the store and buys eggs, also buys milk, Uncle Ben could rearrange his store according to that. By placing the milk next to the eggs, customers buying one of the products and realize they also need/want the other. Also, he could place them at opposite ends of the store so customers must navigate through all the isles and possibly see other items they want.
The most helpful technique for Uncle Bob to use would be the Association rule. This is because he offers a wide array of products. Using the association rule will help Uncle Bob see what his customers buy in association with other products. This can help him place things in certain places throughout the store which would help him sell more products.
Uncle Bob should use association rules. This technique will allow him to find relationships between products. He can find the association between different products such as milk and chips. This will allow him to see if when people buy milk they also buy chips. This will allow Uncle Bob to better place these products around the store. He might want to set the milk on the opposite side of the store so that people have to walk around in order to get to the chips. Using the association rules he could also put them right next to each other. Association rules will help Uncle Bob and his business. If he uses the association rules, he might be able to see what items are bought together.
Market basket analysis assists retailers in understanding the purchase behavior of customers. The information obtained can then be used for purposes of cross-selling and up-selling, in addition to influencing sales promotions, loyalty programs, store design, and discount plans. For example, Home Depot might want to figure out how it will set up its promotions for the holiday and winter season. They don’t want to put on sale two products such gloves, ice scrapers, salt, and shovels if an impending storm is coming along the way because discounting those products simultaneously would yield low profits. The data they would want to collect would be customer purchases. That is, if snow shovels went up, did that drive the sales of any other related product? Customers who bought shovels were they also likely to buy ice scrapers?
I would suggest Uncle Bob to use association rules. It not only help its users to determine association between products, it’s also extremely easy to operate. By getting to know the association between various products, he will know what to order and in what quantity. Also, he will be able to rearrange the store according to the strength of a particular relationship. For example if the association between shoes and socks is close to 1, Uncle Bob can place socks near or in the shoe section. This will help the customers as well as help increase the sales.
One of the most interesting applications of association rules is the Music Genome Project which fuels Pandora Radio. This project is intended to “capture the essence of music at the fundamental level” by categorizing songs as having attributes. There are over 400 attributes that can be associated with a song, ranging from rhythm/meter, structure/composition, tonality, feel, type of instrument/vocals/lyrics, recording techniques, and influences. Based on a complex algorithm which factors these individual attributes, specific combinations among them, and user inputs of “thumbs up” or “thumbs down,” the Music Genome Project can determine a listener’s tastes and generate a radio station targeted specifically to their preferences.
Personally, I am a huge Pandora fan. I have been using it for years and have really “trained” all of my stations to know what I like and what I don’t. If you don’t use it now, start. Pandora knows what you like
I advise Uncle Bob to use association rules. I would advise him in using this because it find relationship between products. It also easy to use so he wouldnt have trouble in actually using this, and he himself would be able to use it without hiring someone to operate it.