What I learned from BI

After taking the BI course this semester, I realized that I have learned a lot of tools, methods, and thoughts to analyse data.

For example, the Gephi part is using for the social network analysis. I downloaded the facebook pattern to the Gephi, and make several changes to see how I am connected with all of my friends. It was really interesting to see the connections between nodes which stand for each of my friend. Basically I have two large groups which are Universito of Arizona and Ball State University. The most interested thing is that these two groups are not connected at all! This means that none of my undergraduate friends come to the UofA like me. Except the Facebook pattern, we also use Gephi to analyse the data from Twitter. Our topic is gas price, so we set the standard keywords like gasprice, gas, oil, petroleum, etc. These gave the visualization of how people tweet with each other. Some people have higher connections because they tweet or get tweet a lot.

The other tool that I am interested in is the OBIEE. I really enjoyed using this tool to analyse the data set. There are so many data for the Cloud Airplane. I actually used different kind of data set to answer the questions asked by our professor. One of the questions was how the delays vary through the whole year or different airports. After I chose and analysed the data, I found out that the delay mostly occurred in December and January. Since they all in the winter, so that might because there are so much snow that caused the delays.

Generally speaking, even though there are so many homework and projects, I did learned a lot from this course. And I believe that all the knowledge and skills I learned from this course can be used someday in the future.

Infographic Resume

The BI homework 6 is to build an infographic resume. Since I want to be a web designer, so I designed a resume as following: 

This was a cover of a magazine named Web Designer. I downloaded the picture and deleted the other features using Photoshop. I added my name and the University of Arizona feature to my resume as well as my experience and education. The iPad thing means that I am able to do all the applications shown on the picture. 

This resume is simple and easy to understand my capability and  working experience. I was thinking about to add some pictures of the website sample, but there is no space to add more. I will probably revise this one when I get plenty of time soon. 

data dimension

In the class today, Dr. Rum has talked about the dimensions of attributes, which are time, membership type, members, items, events, Quantity, sales amount, total numbers of members, while the facts are quantity and sales amount. The quantity has three dimensions: time, items, members, (TSK, ISK, MSK), and the vendor is not part of the schema. Besides, we need to set the schema to validate the slowly changing dimensions, for example, we need to store the history, so that when we retrieve the data, we know that there is something that has changed from previous record. Alternative, we can use to create new tuple or add an attribute to make the data schema more accurate. For example, we can add a new column named "new xxx" while keep the "old xxx" at the same time. So there is no need to add another SK.

Since the dimension is short but fat, so there are so many attributes and we can use the hybrid methods to solve the problem, e.g. creating new column for some attributes, and dealing with the slowing changing dimensions for other attributes. Designing a dimensional model is kind of a art. It comes to the thinking side to decide which attribute to change, and technical side on what tools are we going to use.

Role playing Dimension can be time, because time can be interpreted and separated way. The airline company can use the time to be a role playing dimension and set some sub-attributes like departure time, arrival time, etc. for the time dimension. Besides, there some dimensions that cannot be predicted like quantity in hand. we can use average, maximum, minimum, etc. to represent the data set.

There are many ways to deal with the data. All the decisions are based on the DBA. So think carefully before execute, then come out with the accurate data.

Network Analysis

In today's class, Dr. Ram talked about the network relationships between nodes and edges.

The above diagram demonstrates the relationship between network nodes and edges. These are two different species. The nodes and edges represent different things for different types of network. For example, for the social network, each node represent the a person and the edges represent the relation between nodes and edges. If it is citation network, then nodes represent the papers and the edges represent citations.

Besides, Dr. Ram also talked about the degree centrality, which is defined as the number of links incident upon a node. The term "closeness" is regarded as a measure of how long it will take to spread information from one node to all the other nodes sequentially. Betweenness centrality can be a measure for quantifying the control of a human on the communication between other humans in a social network. Eigenvector centrality is the measure of the influence of a node in a network, which can be assigned scores to all nodes based on the concept that connections to high-scoring nodes contribute more to the score of the node than equal connections to low-scoring nodes.

A clustering coefficient is a measure of degree to which nodes in a graph tend to cluster together. in most real-world social networks, nodes tend to create tightly knit groups characterized by a relatively high density of ties.

The adjacency matrix is a matrix with rows and columns labeled by graph vertices. The rows and columns can be all the nodes. If two nodes have connections, then we can put "1" in it, otherwise put a "0".

In additional to all the knowledge mentioned above, Dr. Ram also introduced the Gephi software which makes a graph that shows how we connect with other people.

This whole concepts are all about the networks between people, which give us a basic information on how to collect and analyze data to make data useful.

Long Tail Keywords

In the BI class on Monday, Dr. Ram has talked about the term "long tail" which means selling small amount of each item with a large amount of different items. Applying the long tail keywords will be pretty powerful for small business, because they are more specific and precise for each item. Let's look at an example first. The following table shows the keywords for "logistics" company. the word phrases has a high competition, and it costs more to show up in the front when people search these keywords. Even though the website shows up, visitors might not specifically want to click this website. The long tail keywords are more specific and contain a lot of words in the phrases. Although the impression will not be increased, there are more people going to click the website once the website show up, which also increases the CTR.

table 1 long tail keywords example

There are so many keywords or phrases we can use for the Adwords, and each of them might only attract few clickers. However, the total clickers will be high as a result of a large amount of the keywords.

"When looking at sales generated by search terms,18.6% of conversionswere from top 10 keywords. Conversely, 81.2% of the conversions were from hundreds of other search terms outside of the top 10. The 80/20 rule works in reverse, providing a sweet spot of opportunity"(Matt Bailey). This is pretty interesting, because when we search the items on eBay or Amazon, we always use more specific keywords instead of only one word. When we did the Adwords research, we found out that even though some keywords like "restaurant", "Indian restaurant" have high impression, the clickers were pretty low which caused low CTR. Thus, when we do our Adwords campaign we cannot just focus on the keywords that has been searched most. Instead of that, we should use the long tail keywords and try to be as specific as we can.

Besides, when talking about the competitors' keywords, we also found out that there are some Chinese restaurants showed up when we search Indian restaurant. However, the CTR was pretty low since almost no body want to go to the Indian website if they are searching Chinese food. For some small companies, who have low differentiation of products might be able to use this strategy. But for our client, Chinese restaurant might be a negative keyword to improve the CTR.

References:

Matt Bailey. Keywords Strategies - The Long Tail. Search Engine Guide. Feb. 20, 2012.

http://www.searchengineguide.com/matt-bailey/keyword-strategies-the-long-tail.php

classmates' presentation

This week, the professor has called some groups to present in front of the class. At the same time, she opened the Twitter wall which can show the tweets of all the class if we use the hash tag "#MIS587SPR ". This was pretty interesting because everyone can see the comments about the presentation immediately. Besides, the presenting team can also answer some questions asked by the audiences.

It was a pity that our group did not get a chance to present that day. However, I learned a lot from others excellent results. Some groups used the Google Analytics to show the data of their client's website and gave the support to their arguments. Most of the teams separate their client's business into to several Adwords groups to make different keywords so the campaign can be more specific and it can attract more visitors using search engine. They also talked about the budget of their three-week campaign like how they are going to spend the $250 each week or per click.

All of the teams were awesome to show their efforts on the online marketing campaign and we definitely can learn from each other to improve our own strategies.

Social Media Network

Since the semester began, the BI class has been more and more interesting to me. This week, Dr. Ram has talked about the impact that social media network had brought to us. To be honest, I have not thought too much about the advantages or usefulness of social media expect making friends or view interesting stuff. However, the truth is, the impact for us is more than that.

Basically speaking, sharing interesting pictures, ideas, videos, etc. with "friends"(can be real-life or virtual) is fun in our daily life. Besides, the diversity of the social media is also a good benefit for us since most of the social media is worldwide. We can know thousands of people from other countries as well as their cultures, languages, or habits.

Furthermore, knowing people who has same working experiences and sharing different thoughts can also improve our insight or comprehension on each event. For people who want to find intern or job, or intend to corporate with other companies, they might want to keep connecting with each other to improve relationship.

Nowadays, college teachers encourage students to use Facebook or Twitters to discuss knowledge that have been taught in class, which is also a good opportunity for students to share their own thoughts about the textbook or class. This is much more efficient than finding articles or thinking about what we learned all by ourselves.

Even though social media has so much benefits, there still are some drawbacks. The most significant one is privacy and security. Being aware of people who pretend to be others to approach you is one of the methods to avoid being hacked.

Finally, I would like to remind myself and others:

we should leverage the time of the social media for best benefit!