Sunday, March 25, 2012

Wolfram Alpha

Wolfram Alpha is a very powerful tool. It mainly differs from Google in the fact that the search engine doesn't try to find webpages. It instead tries to compute answers to certain queries. When you search a term or phrase on Wolfram Alpha, you are not redirected to a bunch of search results containing to links of various pages. Instead, Wolfram computes your query and provides an answer.

For example, I searched "4,000,000 Euros USD" and wolfram was able to convert the currency into US dollars. Pretty neat. There is so much capability with the tool. It will calculate math formulas, stock prices, give detailed company information etc. etc. From my findings, the tool's capability seemed pretty vast.

I really like how query results are displayed on the wolfram alpha site. The tool seems really useful when looking for a specific answer rather than having to search for it like on Google. Because the results are displayed in the native window, navigating Wolfram is really simple.

Overall, I think the tool is great and will use it often

Thursday, March 22, 2012

Twitter Infographic


create infographics with visual.ly The capability that Visual.ly has when it comes to social media data visualization is pretty remarkable. The web app allows for pretty useful analytics of Facebook and Twitter. For my infographic, I explored the hash tag analytic feature. This particular infographic displays how many times "#Kony" was used over the past month. Unless, you were living under a rock, most social media users heard in some form or fashion about the Kony movement. The hashtag "Kony" rapidly trended once a documentary was released on youtube discussing an African warlord named Joseph Kony. The You Tube video received a record amount of hits as well. Using this tool, it's really interesting to see just how quickly and rapidly "#Kony" came onto the social media scene. As the graphic shows, the hashtag peaks with 137,000 posts on its peak day; pretty impressive way to view the social media side of a web trend.

Thursday, March 8, 2012

Twitter Sentiment

Wordle: Untitled

For my first Twitter Sentiment Analysis, I used "Twitter Sentiment". In Recent sports news, Peyton Manning has been all over the news as he has decided to leave the Indianapolis Colts. If you turn on ESPN or any sports news channel, Manning is the main topic of conversation. With that said, a lot of people have different feelings and opinions on Manning's decision. Some are predicting where he will go and some are voicing positive/negative feelings about the change. When "Peyton Manning" was searched using Twitter Sentiment, the sentiment turned up 70% negative. One Twitter user even posted, "Man, this whole Peyton Manning this just really sucks." The word cloud did an ok job at expressing this sentiment. A lot of the larger words including team names that Manning is speculated to be drafted by. My only criticism of Wordle is that there is no way to remove words like "posted 2 minutes/hours/etc. ago" that are common with every tweet.

Wordle: Untitled

For my second Twitter sentiment analysis, I searched "Charter Internet". I have actually negatively tweeted about Charter's unreliable internet service. Within 5 minutes, I was tweeted at by a member of "Charter's Sentiment Response Team" trying to address the issue; somewhat impressed. I used Tweet Tone to conduct a search of Charter Internet. Results were kind of scattered across the board. The top sentiments were content, angry and sad. I feel like more customers were either angry or sad because when combined, those 2 sentiments were more than the content totals. The word cloud was fairly representative of the sentiment analysis as it contained many descriptive words that directly reflect user sentiment. With the word cloud it was interesting to see that many users were tweeting from mobile devices as well as tweeting directly at Charter.

Monday, March 5, 2012

Click Fox




1. Click Fox is a business intelligence tool that helps companies track and improve customer experience. Especially in service industries, customer satisfaction is crucial to profits and market share. Click Fox's "claim to fame" is that it provides a complete "holistic" view of a customers experience. Often many BI tools provide a "silo" view of customer data. With Click Fox, companies can track customer segmentation, buying patterns, where they've shopped, wait times, etc. etc. etc. etc. Many BI tools do this, but Click Fox claims that it truly can make meaningful use of all the data and connect it to help guide decision making that will increase customer retention and profits. Through Click Fox's systems and software, all customer "touch points" are tracked and analyzed. The software presents the customer's experience as a detailed journey and that can be backed by statistical data. At a high level overview, Click Fox provides quantifiable data for a qualitative customer experience with a particular company. Its 3 main areas of focus include: customer retention, operational efficiency, and customer satisfaction.

2. Sprint Nextel is client of Click Fox. Especially in the cellular service industry, the main driver for success is customer satisfaction. In 2008, Dan Hesse took over CEO at Sprint Nextel and recognized many customer service issues needed to be addressed (ex- reduce reasons customers needed to call customer service, solve problems on a first call basis, etc. etc). Click Fox help segment Sprint customers and address needs. Through a top down approach Sprint found various aspects that could be improved in the customer service experience (specifically user phone issues) to increase customer satisfaction (ex- simple tasks being promptly solved by an automated system vs. live agent).Through findings about customer satisfaction and touch points, Sprint Nextel actually saved $45 million dollars on 10 million calls that were able to be contained in its automated interactive voice recognition (IVR) system. In addition, Sprint Nextel actually was awarded Customer Service Awards through Gartner because of their use of Click Fox.

http://www.clickfox.com/blog/sprint-leveraged-cea-earn-top-industry-honors/
http://www.clickfox.com/pdf/2010%20reprint_ClickFox%20for%20sprintnextel.pdf

3. This was the first time that I have ever heard of Click Fox. Through reading the case study on Sprint Nextel and researching, Click Fox really demonstrated how useful customer experience analytics can be. It is an emerging field currently with lots of room to expand. Overall, Click Fox's solutions seemed to be very innovative yet extremely useful for all large businesses.However, it was hard to find screenshots of the software or specifically what the tools tracked and did. Everything I found seemed to be very high-level, and I would like to find out more information on the inner workings of Click Fox's solutions and specifically how they apply to certain businesses.




Hadoop

1. Hadoop is a big deal because it is a system that allows very large amounts of non-relational data to be processed among a "cluster" of machines. Using a technique called "Map Reducing" companies are batch processing large amounts of data (Petrabytes worth of information).Facebook and Yahoo are among some of the top companies that use Hadoop. Hadoop is beneficial to large companies like this because it can process and store large amounts of data, and it is an open source platform. 2. Cloudera is the company founded by Christophe Bisciglia that provides/distributes Hadoop services and software. Companies using Hadoop work with Cloudera to development and contribute source code/improvements hence the open source nature of Hadoop. 3. PIG is a high level programming language developed by Yahoo that allows for easy ad-hoc queries. PIG is a great language for people who aren't extremely code oriented. A simple word count query can essentially be run but over a vast amount of data and machines. 4.HIVE is the data warehouse structure that Hadoop uses to process its data. It was originally developed by Facebook and allows users to summarize data and perform ad-hoc queries. HIVE uses a similar to SQL query language called HiveQL. 5. Cassandra is an Apache system similar to Hadoop. It processes and stores data in large amounts. According to the Apache website, the largest cluster of Cassandra machines processes 300 TB of data on 400 machines. Cassandra is supposedly better at "high-volume real-time transaction processing" than Hadoop according to its website. 6. Mahout is an Apache project that basically exists for users to create, share, and store algorithms that are licensed by Apache. According to Mahout's website, the main goal is to "build scalable learning libraries". Mahout helps supports 4 main use cases: Recommendation mining (finding what a user might like based on previous searches), clustering (examining a batch set of data), classification (assigning unlabeled documentation based of existing documents and data), frequency mining (seeing how often a set of data is queried or processed together).

Monday, February 27, 2012

Many Eyes

This data set from the US Census Bureau reflects employment statistics as well marriage statistics. The data reflects the percentage of individuals who are married based on what working class they fall into: Male/Female employed and Male/Female unemployed. The data is visualized into a heat map with darker brown states indicating a higher percentage of married people for that particular job class. There are certainly many external factors, but I thought it was interesting to see different states and how the 2 things related.

This data set was really neat because it shows internet usage. Within the pie chart you can break it down by various age groups. At the highest level it wasn't surprising to see that emailing, checking news, weather, and social networking were some of the most widely used internet activities. Among smaller usage were things like downloading podcasts or participating in an online auction. The visualization helps recognize trends. For example Females are using social networking sites more than males.

NOTE: For some reason, the Many Eyes site stopped working and publishing my visualizations..even though I could see them. The above link may not work and I had a lot of technical difficulties. This last visualization I had depicted internet usage among various world countries. Not surprisingly, Asia had one of the highest usage rates in the world. The US actually is not a top user. Looking at the population size of each country, it is also interesting to see how that correlates to amount of people using the internet.

Energy Production - comparing the US and China

In recent years, China and the US have rapidly increased their energy production. The graph indicates that China is increasing its energy production but not their GDP capita. The US is increasing both. This could be due to the fact that many areas in China are majorly underdeveloped compared to the US as a whole. This graph depicts the relationship between CO2 emissions and GDP production. One may think that the more a country produced, the more CO2 it would emit due to transportation and manufacturing related to any kind of industrial production. However, looking at the US, the graph reveals a constant increase in GDP production while CO2 emission stays relatively constant. This trend could be related to current CO2 regulations and/or emerging "green" technologies that allow the US to produce at higher levels.