Teresa Ovalle

Welcome to Me

Responses to Week One Discussion (ICM 501)

I chose the following three questions (of 11) to respond to for this week.

3. The University of Minnesota study concluded that some rating scales are more preferred than others. Amazon, Netflix, and other retailers use numbers, stars, likes and other forms of rating scales. Do you feel more confident in a product’s actual rating based on a particular scale like stars versus numbers? What is your ideal rating scale?

My response:

I don’t think stars are more accurate than numbers.  I think it’s a preference.  I personally prefer stars over numbers.  Numbers seems cold, where as the stars seem more personal.  That may not make sense to anyone but me.

In this study, Cosley et al did pointed out that rating system users did prefer a half-star system.  That would be a great system.  Though most systems seem to use a five-star system, a 1 to 5 scale doesn’t cut it. If I had that additional half point option, I would use it.  There is a big difference between the 2 and 3 and 3 and 4 on the scale.  Having the half-point option would offer an additional level of detail.  If the study showed that most raters preferred a half-star scale, then why don’t more recommender systems use one?

8. Fandango is a popular movie ticket website, that I think most of us are familiar with. Not only does Fandango allow you to purchase movie tickets online, they now provide you with both fan reviews and critic reviews. Based on your own experience with watching movies, your knowledge of actors, and own personal like and dislikes, would you take into consideration these reviews when deciding to purchase a ticket? Or do you already have your mind made up when you go to Fandango or other ticketing sites? Would you say Fan Reviews are more credible or Critic Reviews and why?

My response:

I love Fandango.  Although I don’t use it to purchase tickets, I will check the movie ratings before I pay to see a movie.  If critics rate the movie with a MUST GO!, I’ll take the review into consideration.  If the critic reviews are low, then I don’t pay much attention to it.  Critics are paid to critique a movie in detail.  I think they enjoy picking apart a movie, whether it is good or bad.  So if the critics rate the movie high, then it adds to my consideration value.

The fan critics, on the other hand, are what I will use to decide whether to spend $10 on a ticket.  I like to think that the fans rate the movie on what they saw and what they experienced while watching it, more on an emotional appeal than on a technical one.

I also enjoy Fandango’s rating system of Oh NO!, No, so-so, GO and MUST GO!  As a person looking at the reviews, I find that the words have more meaning than a traditional five-star system.  I’m more likely to take the recommendations because of the Fandango rating system than I would otherwise.

9.On Amazon you can buy just about anything. You can also write up a review, give a product a rating and, look – There’s a product you should consider because you just bought whatever it is you just added to your cart! Recently a product review on Amazon went viral. A brand of Sugar-free Gummy Bears sold on Amazon, has pages upon pages of reviews depicting what I’ll call gastrointestinal issues. I’ve never had this brad on Gummy Bears, and after reading these reviews, I probably never will. How much do you take into consideration the reviews of your peers in an online market place? Are they more credible? On the flip side of that, do you think that some of these people are merely piggy-backing on what was already said just to make this situation more comical?

My response:

I’m almost certain I tried the Gummy Bears you are referring to.  My colleague who shared them with me did not warn me of the ‘gastrointestinal issues’ until I emailed to ask her, “What the heck?”  That led to an extremely humorous email exchange.  I will spare you the details.

Movie reviews are one thing.  Product reviews are completely different.  Amazon uses the five-star rating system, as do the movie rating systems, but Amazon also offers the rater an opportunity to write about their experience with the product.  This I find very helpful. 

I start with the five-star ratings.  I skim through them to look for the detailed ratings, the ones with depth.  I spend some time reading them to understand better the specifics as to why the rater liked this particular product.  Then I move to the one- and two-star ratings.  I do the same with these ratings.  I want to learn why the rater disliked the product and, if there was opportunity for resolution of the disliked product or experience, I would want to know how it ended.

There were times that a product was rated high, but the one- and two-star ratings more clearly spelled out as to why I shouldn’t buy the product.  So I didn’t.

I think the Gummy Bear situation may be a separate issue for the raters.  I do think that once more raters caught on to the humor of the issue; it’s likely that more and more raters chimed in to add their own story.

I hope that’s not the case in general.  Though Cosley et al. proved that raters have a tendency to rate similarly to those who rated before them, I’d like to think that most people take the time to rate a product fairly and justly.  

Comments Off on Responses to Week One Discussion (ICM 501)

Netflix, Fandango and the Human Element

In this post, I’ll compare Netflix and Fandango recommender systems that use a five-star rating system.  I’ll also discuss how the human element may play into these systems, and why I think it’s important.

I enjoy sharing my opinion with others.  I’m the person that fills out surveys and tells you what I think.  This is why I find the recommender systems very interesting.

Although Netflix and Fandango are movie-rating services, only Netflix offers a movie service via DVD or streaming. Fandango is only a movie ratings company that doesn’t provide an outside service.  The difference between Netflix and Fandango is important to note because the user interface is a little different.  Instead of rating movies on both services, I can order a recommended movie on Netflix.

It was after reading, “This Psychologist Might Outsmart the Math Brains Competing for the Netflix Prize” by Jordan Ellenberg, that I visited Netflix to take a look at how Netflix allowed me to rate movies as well as how they recommend particular choices to me.

The Netflix system is simple and user friendly.  When I add a new movie to my queue, a screen pops up that offers a number of suggested shows that Netflix thinks I might enjoy.  These shows are predictions.  That is; I’m being offered movies that other users similar to me have already rated.  Netflix is predicting what shows I might enjoy.

I can also pick a genre and rate all the movies that Netflix has prepared for me to rate.  The site states, “The more you rate, the better our suggestions.”  Netflix uses a five-star recommender system but also offers a “not interested” button.  I like this button.  Once I click it the movie will never show up in my predictions again.  As I build my rating numbers, it’s less likely that other similar movies I rate low or I’m not interested in will appear in my list.

Also available via my Netflix account is a Taste Profile.  Under the Taste Profile tab is the Taste Preferences section.  This is a long list of descriptive words such as Absurd, Cynical, Mind-blowing and Visionary.  The user rates the words by the frequency she watches a show that is descriptive to the word.  The user rates the word with Never, Sometimes or Often. I think the more defined options a user has to choose from, the better a rating system will be. I also think that this is where the human element comes into play.  The more choices Netflix offers me to define my preferences, the more likely I will have great movie choices waiting for me in my predictions box.

Fandango, on a different note, uses a five-star rating systems, but instead of clicking on a star, the raters click on the following choices: Oh NO!, No, so-so, GO and MUST GO!  This is a great choice of rating words!  With five simple words or word-phrases, Fandango has added a unique but absolute human element.  There’s no need for Fandango to look deeper.  They are not recommending movies to me.  They are telling me what others thought of a particular movie.

Rather than a three-star rating that offers nothing more than a 3 on the scale of 1 to 5, “so-so” tells me more.  It wasn’t good, but if I enjoy that genre of movie, I may think it’s okay.  So-so speaks volumes.

According to Cosley et al. and their recommender test studies, raters did show a preference for a half-star (0.5 to 5) rating system over a binary scale (thumbs up and thumbs down) and a no-zero scale (-3 to +3 with no zero).  I think fandango’s word-option choices answer that preference.  It doesn’t have the half-star system, but it does allow the user to feel as though she truly has a say in her final answer.

Netflix went a step further.  In 2006, Netflix offered a 1 million dollar prize to anyone who could create a recommender system that could improve its current system by 10 percent.  Netflix had a number of contestants, most of which were mathematicians.  One contestant, however, was a psychologist.  Gavin Potter thought the way to win was with an algorithm that included a human element.

“’The fact that these ratings were made by humans seems to me to be an important piece of information that should be and needs to be used,’ he [Potter] says.”

I agree with Potter.  I think the human element is important, however, I later learned that the Netflix contest winner was not Potter.  Potter came in 17th overall with a best test score of 0.8662 and improvement percentage of 9.06.

The contest winner was BellKor’s Pragmatic Chaos, a group of researchers from AT&T, with the best test score of 0.8567 and improvement percentage of 10.06.

Although the numbers may not lie, there is a human element built into the recommender systems I use most.  Netflix and Fandango both offer me a chance to share my opinion in a fun and easy way.  Both systems also tell me what others think so I can take that into account if I choose to see a movie.  So with the way the recommender systems work for each company, I feel like I’m being heard.

References

1. Cosley, D., Lam, S., Albert, I., Konstan, J., & Riedl, J. (2003). Is seeing believing? How recommender system interfaces affect users’ opinions.

2. Ellenberg, J. (2008, February 25). This psychologist might outsmart the math brains competing for the Netflix Prize.

3. Netflixprize.com

Comments Off on Netflix, Fandango and the Human Element

My Favorite Social Media Platform

I fought for a long time to not have a Facebook account.  I didn’t see the need for it, nor did I want to get involved with it.  I had one friend after another request that I join — but I continually refused.  I ultimately received email requests from four friends in one day.  I’m not certain if they collaborated or if it was coincidence, but that was the last of my resolve.  I finally joined Facebook – and I love it!  It is my favorite social media platform.

When I opened my account, it only took me minutes to figure out why my friends wanted me to join and why I now suggest to family members to join; it’s the best way to stay connected.  I enjoy sharing life in real-time — rather than having to call someone or send them a letter.  I can connect with dozens of people instantly.

I currently manage my Pistol Packing Ladies (PPL) Facebook page as well as my church’s Facebook page.  I do my best to update both pages on a regular, daily basis.

I particularly love using Facebook for my PPL business.  It’s fun, and it’s real.  I can post gun-related issues, motivational and inspirational quotes and photos, and I can update the site with photo albums of our shooting events.  I truly enjoy sharing what we do in our club with the outside world.

For my church page, I enjoy finding a good Bible verse or a Christian quote, picture – something that speaks to me — and post it for others to enjoy, hoping the post makes them feel connected.

I haven’t had any bad experiences with Facebook.  However, the only negative thought that I have — and it’s more of a longing — is my wish that Facebook had been created when I was younger.  I would have enjoyed having Facebook as a young adult.  I could have shared my Marine Corps experience as it happened rather than through ‘snail mail’ and phone calls.  I could have shared moments in time, photos, conversations and a host of other memories that would have been captured forever, if only Facebook had been there.

Facebook has offered one challenge.  Because the PPL page is gun related, I can’t advertise the page.  I was annoyed with this at first.  I wanted to gain a fan base and thought I had to advertise, but I realized that people were still able to find us and ‘like’ us on their own.  They share our page with others, which brings us more ‘likes’.  I’ve realized that I don’t need to advertise.  The people that I want to ‘like’ us do.

Which brings me to “The Cluetrain Manifesto”.  The Cluetrain Manifesto: The End of Business as Usual (2001, Levine, Locke, Searls & Weinberger).  The following passage does a good job of summarizing my thoughts and comments from above.

“The first markets were filled with talk. Some of it was about goods and products. Some of it was news, opinion, and gossip. Little of it mattered to everyone; all of it engaged someone.”

The first item mentioned in ’95 Theses’ is Markets are conversations.  Facebook is an international market.  It allows millions of people to visit the market of their choice every day.  They have conversations with people about issues and things and ideas with other people from around the world.  Everyone in any given market is speaking the same language, the language of a particular interest.  People want to talk to other people about what interests them, whether a product, news, opinion or gossip.

People want to speak with, be heard by and be spoken to by a person.  The markets are now open.  Let’s go visit.

Comments Off on My Favorite Social Media Platform

A Great Grammar Review for Everyone

A good review for anyone who is interested.

http://sparkcharts.sparknotes.com/writing/englishgrammar/section1.php

Comments Off on A Great Grammar Review for Everyone

Conformity – Simply Stated

How interesting to learn that behavioral sciences are interlaced with social media algorithms to the degree in which they are. The human dimension will challenge even the best algorithm.

When I finished watching this week’s lecture on Ratings and Recommendations, I got online and googled ‘social conformity algorithm’ to see what popped up. There were quite a few links that looked interesting, but the one titled, “What is Conformity?”, caught my eye. It’s a short article written by Sam McLeod, a psychology lecturer at Wigan and Leigh College in the U.K.

McLeod states that conformity, “is a type of social influence involving a change in belief or behavior in order to fit in with a group.” He then adds that, “The term conformity is often used to indicate an agreement to the majority position, brought about either by a desire to ‘fit in’ or be liked (normative) or because of a desire to be correct (informational), or simply to conform to a social role (identification).”

Note that Kelman (1958) distinguished three categories of conformity as compliance, internalization and identification. Man (1969) distinguished three different types of conformity as normative, informative and ingratiational.

If you take a few moments to read the article in its entirety, you’ll quickly realize it’s a simply stated psychology paper, yet it ties in to the issues we are reading about this week.

http://www.simplypsychology.org/conformity.html

Think about what we learned in the lecture about recommendations and how most people won’t rate what how they truly feel about an issue, a subject, a move, etc., particularly if there are other ratings already present. That’s conformity.

Let’s use an online movie rater as an example.

A movie rater decides to rate a movie on Netflix and thinks to himself, “Hmm, I think I’ll give this movie four stars.”  Of the six conformity choices we have to choose from which is he?

He’s unlikely to be ‘normative’. That is, there is no group pressure. The person rating the movie can choose what he wants without discourse. I also doubt that it’s ‘compliance’. There’s no reason for the rater to disagree with anyone – he’s online.

It’s doubtful that he is ‘ingratiational’. He may be the kind of person who wants to impress others, but again, he’s online. There’s no one to impress.

The rater could be ‘informational’. Perhaps he’s rating a movie just to rate a movie, but really doesn’t know much about it. I’ve done that. I think I’ve seen it. I kind of remember it being good, so why not rate it, right?

He could be an ‘internalization’ rater. However, I think that’s unlikely.  Because he’s online, he has a certain level of anonymity, so why worry about what others think? There’s no rational reason to try to fit in with others.

So the best choice we have is for him to be an ‘identification’ rater. He agrees with the other ratings he sees, but doesn’t have to change his personal opinion.

This was a simple experiment to apply some of what I’m learning to a process that makes sense to me.  I think it proves that the human dimension can not be categorized into simple boxes with suggestive elements. The human dimension changes everything, to include algorithms.

Comments Off on Conformity – Simply Stated

ICM 522 Intro

This is my intro video for ICM 522.  It took several recording to get it right, but it’s done.  I’m certain I will be a master at Screencast by the time I’m done with this course.  :-)  Take a look and tell me what you think.

Comments Off on ICM 522 Intro

Screencast-o-matic Success!

It took some trial and error and a few tutorials, but by gosh, I think I’ve got!  I’m sure there are many things to still learn, but the basic mechanics are easy to use.  I decided to use my PPL website as a demo.  It worked very nicely.

Comments Off on Screencast-o-matic Success!

Screencast-o-matic Experiment

I’ve been experimenting with Screencast-o-matic today.  My first experiment was not a fun one, but that’s a part of learning.  I did manage to finish and upload it without difficulty.  I’m very pleased with myself.

Comments Off on Screencast-o-matic Experiment

Social Media Sites

To get ready for the onslaught of school work undoubtedly coming my way in just a few days, I decided to set up social media accounts that I think I’ll need.  It’s going to be very cool to see how the different platforms work and what people use them for.  I’m already forming opinions about a couple of them, but I’m not going to share my thoughts until I learn more.

Here is my list of sites:

Facebook: Teresao4501

Twitter: Teresao4501

Pinterest: Teresao4501

Google+: Teresao4501

Comments Off on Social Media Sites

Changing my direction

After consulting with a classmate in my QU program and learning more about what a blog is,  I decided to change the direction of my blog and focus on my QU experience and capture as many of life’s moments as I can along the way.   This includes writing about things I learn, things I experience while going through the program and how this program has enhanced my professional and personal knowledge, skills and attitude toward interactive media.

I’m certain I will venture off the path from time to time, but that is to be expected.  Life happens, but I will do my best to keep the course.

I hope you’ll join me for what I am sure will be an amazing adventure over the next two years.

Hang on!  Here we go…

Comments Off on Changing my direction