sidradis2000-blog

[sidradis2000-blog]

IS2000 Blog: Week 12

Link of the week: https://des.wa.gov/services/risk-management/about-risk-management/enterprise-risk-management/root-cause-analysis

During this week’s classes we learned about information-based processes and process analytics. Here we talked about creating flowcharts to organize processes and we learned about computing total process times. We also learned about how to determine the time or cost of a project from estimations from a bunch of employees using a method called wide banned Delphi. And while we didn’t discuss this in class, we also went over Little’s Law which determines the long term average number of customers.

When we went over flowcharts I originally was under the assumption that it would be an easy concept to go over. We had went over UML diagrams and Ontologies, so I figured I would have an easy time with the concept and assignment. But trying to figure out how to organize the flowcharts so they were readable and trying to implement swimlanes had just made the process more convoluted and time consuming. Also when interpreting the DNS forward lookup process, or the process that gets the ip of the website using a domain name, is quite hard to turn into a flowchart especially with the concept of recursive querying. Recursive querying is essentially a kind of query in which do a the job of fetching an answer but during the process, response might also be another query, so you continue until you get the answer.

In class we discussed how to calculate the completion time of a flowchart. One thing that never really occurred to me until we had to implement it into an assignment was that if you have a parallel process, to total time would simply be the maximum time between each of the branches. When completing the assignment I also had to figure out how to implement expected values in determining the total completion time of a process, so you multiply the probability of the event times the actually time the process takes.

Another key concept that we went over is the wide banned Delphi. This is a process in which you gather estimates about a project detail from various professionals. You need to evaluate their optimistic, pessimistic, and best response. This process is very useful because you evaluate a large set of estimations to make an accurate prediction. I am in biostats, so we had done a similar sort of process when evaluating mean of different collected samples and constructing confidence intervals to account for variation.

While we didn’t cover this concept in class, I found Little’s Law to be very interesting. It essentially states that the long term average number of customers is determined by multiplying the average waiting time to the arrival rate of customers. As mentioned in the required reading, I found this law to be very useful because it helps gives startups a helpful gauge when discussing how to increase consumer/customer base.

Another process based tool that I learned about in my CS Overview class is an RCA, or a root cause analysis. This is essentially the process of organizing the series of events that cause a certain phenomenons, and accessing how each of those events occurred. Linked above is an article one Root Cause Analysis. I find this to be a very important process, even though it is timely and costly, because it helps you not only determine the root cause of a problem, but helps you create a solution/solutions to prevent similar issues from occurring again.

[sidradis2000-blog]

IS2000 Blogs: Week 11

Link of the week: https://www.techworld.com/picture-gallery/security/10-biometric-technologies-that-will-kill-passwords-from-fingerprints-iris-scans-selfies-3637418/

During this week’s lectures, we discussed about Information Security, Privacy and Confidentiality. We covered a variety of topics such as networks to data collection methods at Facebook and Google. We discussed this topic in another class but the concept of business ethics is also important here as well. One of the major issues that occurred that was relevant to the discussion was the whole Facebook-Cambridge Analytica event and it is through this we discussed the previously mentioned topics and their implications. This was one of my favorite classes because it wasn’t very information heavy and the topics we discussed were very interesting.

One of the first topics we discussed in class was about networks and servers. We discussed about how people can privately view the web and how information can be securely sent throughout the web using the SSL protocol (Secure Socket Layer). We also discussed about VPNs or virtual private networks and their implications regarding Information Security. The lecture on vpns and vps’s had reminded me of a time back in highschool where my friends and I had made a VPS in order to view social media in school.

Our school system had implemented a unreasonably strong internet security and it prevented us from viewing a variety of webpages. One of the blacklisted keywords was “gun” and for a U.S. History project we were not able to look up information regarding the second amendment. So to bypass this, we built a vps using a raspberry pi and set it up at a friends house. We then connected to it on the school wifi and used that as our mediator to browse the internet.

When discussing information privacy, we discussed the essentials of business ethics and about the Cambridge Analytica Scandal. We talked about the audiences or the focuses of buisnesses are either for maximizing profits, to benefit the shareholders, or benefit the stakeholders, which include users/customers, employees, shareholders. The activity we had done in class was based off the Equifax breach and one classmate had to decide whether they don’t inform their customers of the hack until they find out where the breach in the system was or if they inform their customers as soon as possible and risk other hackers exploiting the breach. While I feel it would have been smart for Equifax to have informed their customers first and help them through the process of securing their information, they instead decided to keep quiet for 3 or so months. This delay is probably why they experience the most backlash. Also probably because they had said customers had to pay them to freeze their credit accounts, even thought Equifax was at fault.

One cool topic that we went over in class was regarding Security and Confidentiality and how to implement authentication systems through biometrics. The article I found above discusses some of the cool biometric technology such as fingerprints and iris scanning. One of the cool biometrics I found was keystrokes, and how computers can determine if the user is and authorized user based on how they type. Apparently typing rhythm is unique per person and is almost impossible to spoof.

[sidradis2000-blog]

IS2000 Blog: Week 9

Link of the week: http://articles.chicagotribune.com/2012-08-07/business/sns-201208071730--tms--savingsgctnzy-a20120807-20120807_1_casino-owner-gamblers-lottery-tickets

Decision Trees are a very effective way for companies and organizations to assess risks and make decisions. It takes into account the different factors surrounding a decisions and make the best decisions for maximizing profit/minimizing costs. From the practice in class to the assignments, we had constructed various decision trees to help these “companies maximize profits. We started the discussions with probabilities and the 4 main ways of deriving these probabilities: subjective, logical, empirical, bayesian. We then used the following understandings of probability to construct decision trees and derive the best decisions.

Logically deduced probabilities are simply probabilities that are determined by the number of desired outcomes over the total number of outcomes. So the odds of rolling a six or flipping heads are all logically deduced probabilities. While I find these deductions to make sense and are intuitive, some events explained through logical probability simply don’t seem right. For example, what is the logical probability of it raining tomorrow? In fact the answer is fifty fifty, either its raining or it’s not raining. Besides this puzzling statement and similar situations, I would say logical probabilities are probably the most reliable probabilities, since they are derived directly by the set outcomes.

Another method of determining probabilities are through subjectivity. If you want to know the chance of something happening and you use your opinion or your intuition, then you now have a subjective probability. These are pretty useful when you want a rough idea about the chance of a certain event occuring. Perhaps you ask a co-worker what the chance of a client purchasing a product again is, and his response would probably be a subjective probability. These aren’t very reliable and probably should be used as gauge or just to give you a general idea.

Empirical Probabilities are probably the most popular and are used the most often. These probabilities are determined by historic data or data collected, so they are based on hard evidence and fact. Such probabilities could be determined after an analyst collected data from surveys and provided the percentage of people who prefer television over the radio or perhaps after a company looks at the prior sale history with a client and predicts the chance the client would want another sale this year. I find these probabilities the most realistic to follow, since they are predictive based on established evidence.

At first I was skeptical of the concept of a decision tree. I was not sure how I would use probabilities to determine a profit maximizing decision. However after we went through a couple examples and learned about Expected values, then the whole decision tree makes sense. Expected values are determined by multiplying the probability times the number of times an event occurs. This is important when determining the decision with the least costs or most profit because if for example if you decide to either park in a garage and pay 15 dollars for a parking fee, or park on the street where there is a 50 percent chance you pay 20 dollars or you pay 0. You might think you pick the garage since its the least risk, but after for accounting for expected values you see that the actual cost of parking on the street is actually 10 dollars. The link above defines Expected value and shows the use case of expected values, through it is gambling, it is still an informative article.

[sidradis2000-blog]

IS 2000 Blogs: Week 8

Link of the week: https://www.tableau.com/

During this weeks lecture, we learned more about XML, SQL, CSV (comma separated value) files, how to use the R programming language to collect the data stored in those data storage formats. The R programming language will basically read these files and organize and store them in a data type known as a data frame.

In the past I have worked with CSV’s as means of data transmission. For robotics we need a develop a system where our “scouters” could record stats on robots on the field via tablets and send that data to a central computer so we can rank the robots according to our needs. So we had the tablet apps generate a qr code that contained a CSV string and the computer have run a python script to use a camera to read the qr code and convert csv to data that can be added to an excel table.

Working with R made me realize how much more data analytics could be done on top of the basic aggregates or built in functions of XML, Excel, or SQL. While Excel has a lot of built in functions, both XML and SQL are very limited, with basic functions like count or average. R however, has many built in functions and also allows the user to add his own functions. Similar to SQL you can get specific “views” of your data using the dollar sign or the aggregate function. R simply provides a extensive array of data analytics that the other 3 can’t provides. I would say the other three provide a great data storage and organization but if you want to get any meaningful information from the data, then you will definitely need to use R.

[sidradis2000-blog]

IS 2000 Blogs: Week 7

Link of the week: https://www.upwork.com/hiring/data/sql-vs-nosql-databases-whats-the-difference/

During this week’s class, we learned learned more about Relational Database systems and a database computer language called SQL, which stands for structured query language. From what we learned about on Wednesday’s class, SQL is a very useful tools for companies and allows for an organized way to store data. While we did not learn all the ins and outs of SQL, we specifically learned about using SQL as an information Retrieval tool.

The more we talked about Relational Database systems, the more I see its usefulness. Working with the Entity relationship diagrams helped me visualize how these database systems function and working with an actual relational database language showed me the practical usage of the language. I also understand the use of Primary and Foriegn keys, and how they proivde the relationships between tables.

While working with SQL, it felt like it was quite similar to XPath, in the sense that you had to specify paths within the data structure to view the desired information. However, SQL seems a lot more useful since you are not only able to access specific information from tables but you can input functions into your SQL query so you can collect computed information, instead of having to collect information from an XML using XPath then using some other program, perhaps an R program, to compute information from the XML.

One thing I wished we could have done with SQL was if we had an assignment similar to our Assignment 1 and assignment 4, where we had to work in a corporate setting and complete a task. I feel that this would given me an even more thorough understanding of the practical uses of SQL. I had a similar thought when working with XML and XSLT assignments, because while I can see the usefulness of these languages, I don’t really get the impression of the impact they have in a real life application.

I have had some prior experience with SQL through work experience. I was in a project where we had to collect information on electricity usage and we had to store the data onto SQL tables. We had built programs that would query these tables and render graphs of real-time electricity usage. While I had overseen the project, and developed the systems to collect the data, I did not actually participate in the data visualization, or the sql querying. I was wondering how these tables had worked or how the programmers queried this data but after this week’s class I understand they were able to aggregate the data we wanted to visualize using the AGGREGATE, WHERE, GROUP BY, and various other clauses.

While looking up information regarding SQL and relational Database system, I found an interesting type of database technology called non-relational database systems, or NoSQL. These NoSQL systems are very reliable when you don’t have structured data or aren’t able to develop a relational database system with a clearly defined schema. They are document oriented, where each document is easy to find but there is no defined structure or organized system.

[sidradis2000-blog]

IS 2000 Blogs: Week 6

Link of the Week: https://incrementaldevelopment.com/xsltrick/

For this week's classes we learned about how to use Xslt and the basic concepts of database and database management. From the recorded lectures, we learned about the basics of sql and the concepts of ERD and functionality of a RDB.

For me, learning XSLT was kind of a challenging. After learning about XML and DTDs, (and the basics of schemas), it was hard to wrap my head around the concept. So an XSLT is an XML that Used to transform other XML’s into other XMLs???. I realized that the mind set that I should have when creating and XSLT file is to just maintain focus on 2 files, the input and the output. An XSLT is just the output XML except the location in the code for actual data are replaced by XSL syntax and XPath references to the input file. After somehow managing to grasp the concepts, figuring out how to implement the knowledge into Assignment 6 was pretty easy. While I still had to constantly refer to w3schools for the XSLT syntax, I still manage to figure out how to complete the assignment.

When I was learning about Entity Relationship Diagrams, I had this deja-vu moment because it was quite similar to UML Diagrams and Ontologies.

[sidradis2000-blog]

Weekly Blog Post: Week 5

Link of the week https://www.json.org/xml.html

During this weeks classes we learned about xml files, DTDs, a little bit about schemas, and X-path. As a continuation of ontologies and structuring our data, XML seems very useful for organizing and categorizing data, since its organized in a tree like structure. Using XML is very useful since you have a text based method for structuring data and for relaying information to readers. I personally found learning about XMLs to be very useful since it allows us to translate the convoluted structures of an ontology into a more presentable form.

One issue, or concern I had with XMLS were that they seemed very rigid, it didn’t really make sense how they would be use for having back and forth communications, but that was resolved when we learned about DTDs. These structures are vital because they set up the structure for storing data into the DTD. Without these structures it seems useless about how we would be able to transfer this information. I find it as if XML files would simply be useless if you did not have the corresponding DTD because it seems it would be virtually impossible off you wanted to collect any information from an XML.

X-Path also seems very useful since this allows use to traverse the XML trees. Both the absolute and the relative methods of traversing a tree seem vital since they allow use to reference data from virtually anywhere on the file. We also learned about how X-path could be used as a translator because it allows us to collect information we desire from files and input them into other functions or programs that only accept a certain file type. This is obviously very useful but each X-path code would be useful for Niche scenarios.

[sidradis2000-blog]

IS 2000 Blogs: Week 3

Blog link of the week: uhsse.org

As the weeks go on, I start to see more of information science in the world around me. Especially after this weeks lecture on information modeling and ontologies, I notice and understand why a particular website is built the way it is or why a piece of software is designed in a specific way. Not only do I have this “newfound perception”, I also start to admire and/or critique website and software designs because I understand their use cases and how they were/were not being followed. The other day, I was admiring the Snapchat app and how effective and simple it use to use. With just 3 main panels, or pages, you are able to fully connect your self with your family and friends.

However I also used an app called PoshMark, where people post apparel they want to sell and I realized how cluttered the app was. It was clear that the app was supposed to let you see items that would fit you, such as shoes that were size 11 or a shirt that was “medium”, but the app would disregard those requirements and continuously show you newly posted items, or items that were popular. It also felt like was trying to implement a chat system or some form of social media but it would basically just be spam since people would just send messages requesting you buy their product.

When I was learning about ontologies,  how to establish your domain, and identify your classes, I found myself reminiscing about my high school coding class where we learned Java. It quite peculiar because the same terminology, or similar terminology, was used when modeling object oriented functions. When we were drawing the diagrams for our sample clothing website in class, I felt like I could visualize the code that would correlate with the UML diagram and how the attributes would represent the field variables.

While I found this week’s lectures on information modeling and ontologies to be very interesting, I was already somewhat familiar with the topic because of prior high school projects. Linked above is a website for my school that I had helped design. I was not explicitly part of this team,  who used the website as their senior project, but I had helped them design an ontology to build the website around where it would highlight important categories for students. The main issue with the prior website was it was cluttered with junk that students didn’t need. It was a hassle for us students to use the website as a resource, even though it contained a lot of valuable information.

We had decided to establish the domain of the website to be a central hub that students and parents could use for all school related matters. We also designed a back end where each teacher had specific portions on the website they could control in-order to display valuable information to their specific students, for example details regarding a summer project. After learning about ontologies and information modeling in class, I look back and see some changes that could be made to make the website more appealing. For examples maybe a different color scheme, or a different layout to display the different pages. Regardless, I still feel like the website accomplishes its “use cases” and is good example of information science at work.

[sidradis2000-blog]

IS2000 Blog: Week 2

Link of the week: https://cyfar.org/ilm_4_5c

During this week’s classes we discussed 2 different topics: Information collection and Information analysis. For information collection we discussed, all the different techniques businesses use to collect information for a project or task. Some examples include Interviews, Focus Groups, Observations, and Surveys. In tangent we discussed about biases which may skew the results of information collected, such as observational selection, confirmation bias, etc. When discussing Information Analysis, we talked about how information is organized, structured, and labeled. We discussed the many use cases for information architecture and the concept of “Information ecology” and the hierarchy of information.

When diving into information architecture, I didn't really know what to expect. I figured we would discussed about how information is organized but I didn’t expect much else. I found the whole concept of the Information Ecology very interesting. It is essentially a venn diagram of the 3 categories which helps organize and label information. The categories are: Users, Content, and Context, where the users references the audience, the content references the actual information, and the context represents the objective of the information. I found that this methodology of categorizing information seems very useful since it doesn’t represents any arbitrary label, but instead organizing the information by its usefulness.

When discussing Information Collection we learned about a variety of techniques that businesses use but one problem I found with almost every one of the techniques are that they are all so resource intensive. They all provide valuable information but it’s unfortunate because a conservative or small businesses would be at a loss since they would be less inclined to spend as much money. One of the information techniques that was my favorite, both in practice and to create for the assignment, was surveys. Surveys not only are the cheapest of the techniques, they can ask a variety of questions for either quantitative or qualitative collection and they are relatively anonymous.

In tangent with Information Collection, we learned about the associated biases. These biases can affect the validity of information and they are very hard to prevent. When going through the list I had a sort of dejavu moment because I could remember the times I myself were affected by these biases. One notable example was when my parents and I experience observational selection bias when we bought a new car. For a couple months, everytime we would drive on the roads we would always get mad about how we see the same exact model on the road. We would play games like counting the number of times we would see the car. Quite ironically it wasn’t that a bunch of people happen to get the car at the same time, but we tend to notice the cars more and wrongly assume the was an increase in frequency.

When looking for information techniques (for the assignment) I ran across an interesting term I didn’t think i would hear for a long time: Personal Digital Assistant or a PDA. As the site linked about says, it is a handheld device that could be used to record data onsite and then uploaded to a computer. While now this technology is outdated and frankly is not user-friendly, I feel that for its time, It was probably very useful, since you now didn’t have to carry a bunch of notebooks or “paper hardware” to record data, and instead use this “Small” and “ergonomic” device.

[sidradis2000-blog]

IS 2000 Blogs: Week 1

Blog Link of the Week: https://www.huffingtonpost.com/entry/what-machine-learning-means-for-the-future-of-data_us_5936e418e4b0cca4f42d9dc4

My first couple classes of IS 2000 have been very informative in regards to defining Information Science and many key concepts as well as discussing the types of work information scientists do. Before I watched the first lecture, I wasn’t really aware of the impacts of an information scientist or the field of information science; it was something I simply took for granted. I also really couldn’t distinguish between a Data Scientist or an Information Scientist.    For the first week of class, it felt as if the first lecture formed a lot of questions and stigmas for information science. I couldn’t really wrap my head around the concept of a job as an information scientist because of how far information science branches out into other disciplines, like philosophy, business, cognitive science, etc. I also questioned how different my major, of Data science, was to Information science. From the definition of information science to many of the careers covered, I had originally thought those were positions of a data scientist. However, it was the second class that had cleared and answered many of my stigmas and questions. I finally realized the clear distinction between the information scientist, who gives meaning to data and set up the requirements for many operations and projects, while its the job of the data scientist to collect that data for the information scientist.    The Huffington Post article linked above had also sparked my curiosity of what contrasts both fields. I read the article after our first lecture, where we defined information science and discussed its implications, and I wondered why the field of Data science was so ingrained in concepts of Artificial Intelligence and Machine Learning, but Information Science is almost never mentioned when discussing those two topics. However, after our lecture on Thursday, I actually saw the connection between Data and Information Science. It was the Data Scientists that would develop the Machine Learning Algorithms, but it was the Information Scientists that would set the criteria for these algorithms and its implications. It was also Information Science that would make judgment calls based off of the data collected by the Data Scientists or determine the ethics of the code or algorithm.    In addition to the many of the realizations I made about Data and Information Science, I also found many of the topics regarding information and knowledge to be interesting. One thing I found very helpful was the distinction between data and information, and how information simply gives purpose, or direction to data. I also found the SECI model and the conversion of knowledge from tacit and explicit to be very reminiscent of my robotics team in high school, such as how our teams mentors would teach the students many tips and tricks in the build room, which symbolizes Socializing (tacit to tacit knowledge) or how our programming team wrote an API handbook for coding the robot with java, which symbolizes Externalization(or tacit to explicit). I am currently reading about and understanding the different ways of collecting information and I am excited to see what the upcoming classes and readings have in store.