In reading, “The Guiding Principles for findable, accessible, interoperable and reusable data” I was able to learn about how to make data available and how to implement the use of data fairly. I believe that the information put forth by this article is incredibly important should be seen as a way to combat the rise of “fake news” in our current social and political spheres. The article stresses that data should be findable, accessible, interoperable and reusable to all in order for it to be considered fair data. Although I agree completely, I cannot help but think about top secret documents and the information they hold and wonder if, because their data isn’t accessible to the greater population, if this data is considered to be illegitimate?
My fellow classmate RC comments, “ Historically, data is often tightly held and those who create data can be biased in their creation and analysis so opening up data flows will create more equitable data”. Though I agree with this statement, could we not hope that the CIA officer dealing with top secret documents would better analyze the data than I would? Because I do not have access to it does it make it that much more likely to be open to biases?  

This article was really interesting and got me thinking a lot about credit, especially based on the interviewees within our data. I wondered if any of them knew exactly what type of impact their 20 minutes interviews were going to make. We have been able to discover so much about the past with just some simple questions being answered about these peoples past. I would love to be able to show the interviewees what we have done with their info, and how useful and nice it was of them to participate in the interviews. I know the LA Museum has used the interviews for even more, so it just shows how much of an impact they have made. The author of Questioning Scholarship: The Digital Humanities highlighted an interesting point from the reading that it is likely within the future more credit will be given on online scholarly articles, so to develop that concept faster we should make sure to give a lot of credit to the interviewees.

The reading was a thread on twitter that highlighted the use and misuse of social media data. It was a very insightful text. I especially liked that the author addressed what runs through peoples minds regarding anonymous release of this data but in fact by releasing different pieces of private information, they give the public the tools to trace this back to its origin.  I follow MG’s thought process on whether or not making accounts “private” could change the amount of information gathered or if the user has no control at all over any of this. 

This article caused me to think significantly about the way I present data in my projects and also what data is accessible to me. It also made me think about the way that data has historically been inaccessible to some groups. In my Environmental Science class we recently watched a movie about the Love Canal, an environmental disaster that was wrought by the lack of information about a certain community’s exposure to chemicals. Not only were the residents unaware of the fact that they were exposed to a significant amount of chemicals, but when they sought out information on their exposure, the data was very difficult to read because of the density and use of scientific language. The data in this situation clearly went against the FAIR guidelines because it was not accessible. As Michael commented, the FAIR guidelines help ensure “effective research and future studies.” In the case of the Love Canal, the data was unable to be effective or influence future studies until it was made more accessible by a scientist who translated it into something more understandable by members outside of the scientific community. This article made me think immediately of the Love Canal situation and made me realize the importance of making all data FAIR. 

This article about domain literacy explains the importance of knowledge about domains, and where we are putting our work and personal information. As we move further into the age of internet and technology, the author wants to stress the importance of knowing simple domain literacy. The author lists terms/topics for which every person who uses the internet should know about. For example, everyone should know what a domain is, how it works, and who can control it. Many individuals as well as companies own their own domains and can pull information from its users. For protection and privacy purposes, it is imperative that users know what information can be taken.  The author states, “You don’t walk into shady establishments in the physical world, and hand over your private information to people you don’t know or trust–we just need to help make people more of aware of the details of doing this in the digital, as well as our physical worlds.”  The purpose of the article is to explain the importance of protecting our own information on the web.

This is a very important topic that everyone who uses the internet should become familiar with, especially as we move farther in the world of technology. 

There were no other posts to comment on regarding this article.

Bing Liu’s article,  Sentiment Analysis and Subjectivity, explains the history behind opinionated text and how it is researched today. He opens his article by defining textual information into two mains types: facts and opinions. Facts are objective, and opinions are subjective expressions that often describe people’s sentiments or feelings towards  an event. Liu explains that throughout history, most of the textual information people have researched have been fact based. Why? Because before the internet, there were few places to publish opinionated text. Before the internet, opinions were often made to friends, family, etc, and were not published. Today, opinions flood the internet. With websites like Amazon and Ebay where consumers can comment on products, and applications like Instagram and Facebook where users can comment on posts, opinions  are constantly being made. This is where sentiment analysis is incredibly effective. Sentiment Analysis takes all the words in a text, determines if the words have a positive or negative sentiment, then returns a “sentiment score” allowing the user of the program to determine whether the overall text is positive or negative. 

Sentiment Analysis is not a full-proof way to tell whether a text is positive or negative as it is susceptible to errors, but in many cases, it is very effective in taking large bodies of information and determining whether it is positive or negative. One of my classmates made an interesting statement about how the computer can determine whether something is positive or negative. BM states, “computer will never understand the emotional values and ever-changing expressions of human beings.” How is it that a computer can determine whether something is positive or negative without emotional values or feelings? Overall, sentiment analysis is an incredibly fascinating topic and has the opportunity to be a very effective strategy for analyzing data in the future.

In reading “What you can, can’t and shouldn’t do with social media data” I found it particularly interesting how data scientists are discovering the relationship between social media and political affiliation for example.  While I believe that the abundance and availability of social media data will provide researchers with valuable insights on these questions, I also believe that the sheer magnitude of data will just confirm phenomena that we already know.  In other words I believe that social media creates more data points that follow a curve that we already plotted. In terms of political affiliation, researchers were well informed about geography and social frameworks that surround each political party affiliation.  

What I find much more interesting is not the relationship between social media and politics is how social media creates politics.  The social media platforms that have popped up are major rivers for the flow of information, ideas, and political rhetoric. With this platform we see the emergence of niche political groups and people rallying around certain ideologies.  Additionally, these social media platforms have created a completely new form of political debate online.

Finally, I think that social media has reinforced confirmation biases within politics.  Coming full circle, one way in which social media data has confirmed our understanding of political party is confirmation bias and friend groups.  A map of facebook friends and their corresponding political ideology show little overlap across party lines. This confirms our understanding of politics and further reinforces a political dichotomy.   

I thought the article was a great overview of the topics of social media data gathering. There are tons of benefits including ease of collection and lack of the Hawthorne effect where people in studies change their behavior merely on the fact that they’re being watched. These benefits make social media data extremely easy to gather and smart knowing that the data is genuine. However, there is also some drawbacks including biased data because of the relative similarity between social media users, these similarities even draw closer depending on which network the data is being gathered from. Facebook users will be much older than snapchat users and there will be many more women on pinterest and men on reddit. Depending on the type of population you are trying to gather data from, this can be a good or bad thing for an analyst.

I think social media is a great way to gather data from a wide range of subjects. Obviously it depends on the type of data you are gathering and the type of people you wish to gather data from, but the ease of collection and genuine answers make social media a great place to capture data for any purpose. Classmate HC agrees that “This gives a newfound accuracy to the data we collect through this medium.” There is a massive amount of accurate data around now that we would have never had access to before social media.

This post made me immediately think of Donald Trump’s tweets. I think that this day in age, especially with Trump, social media has surpassed being simply a platform of socialization but one that was almost likened to that of a newspaper. For instance, Trump tweets his every move. If social media were more privatized and protected, it would be impossible for reporters and other nations to access the information they needed. I don’t necessarily think that Twitter is the appropriate platform to publish national decisions, but I appreciate its accessibility.

One question I had while reading this article was how different levels of privacy related to how much of one’s data researchers were able to access. For instance, whether one’s instagram account was public or private. Was research based primarily off accounts that were public? Or do the terms and conditions give permission to some researchers regardless of the privacy settings of the account?

I also disagreed with the point made about how the ability to conduct research without direct consent allowed researchers to avoid the Bradley effect, Hawthorne effect, and response bias. I think that people cultivate their online personas, and while they might not be cultivated with the intention of responding to certain research questions, they are not truly authentic.

I did not see any other annotations to comment on.

One of the many benefits of social media data which was mentioned in the text (What you can, can’t and shouldn’t do with social media data) by Rachael taxman was how it made making specific studies and statistics much more manageable. What used to take years and sometimes decades can now be done in days and with a fraction of the number of workers. The benefits of social media data are more; it turned out that because of physiological reasons the data collected from social media platforms is more honest and accurate. That is because the information is obtained without any direct contact. Turning to the things that you cannot get from social media data, in the text it mentions that the data collected from the social media will defiantly be biased since you are just examining a particular group and not the whole population of a place. Also, you do not know about whom you are collecting data from. Moreover, there are certain privacy agreements that you have to abide that restricts you. And in the end of course it talked about the ethical aspect of collecting data from social media sites which were mainly about making sure that who ever’s data was taken it should be under all the ethical rules. 

I think I agree with HC opinion on how “this should just be another tool researchers use, and not replace all forms of data collection.”