Author Archives: Ryan Yoakum

Unreadable: Challenges and Critical Pedagogy to Optical Character Recognition Software 

Posted on by
Reply

In the 21st century, Optical Character Recognition (OCR) software has fundamentally changed how we search for information. OCR is the process of taking images with text and making them searchable. The implications of OCR vary from allowing searchability on massive databases to promoting accessibility by making screen readers a possibility. While this is all incredibly helpful, it is not without fault, as there are still many challenges to the OCR process that create barriers for certain projects. There are also some natural limitations to using this software that especially have consequences for time-sensitive projects, but other factors within human control have negatively influenced the development of OCR technology in general. This blog post will explore two issues: the amount of human labor required on an OCR project and the Western biases of this kind of software. 

Some text in ABBYY FineReader. Not all of the appropriate text is contained within a box, indicating the human labor that needs to go in to correct this.
Public Domain Image

Human Labor Requirements 

While OCR can save an incredible amount of time, it is not a completely automated system. For printed documents from the 20th-21st century, most programs can guarantee a 95-99% accuracy rate. The same is not true, however, for older documents. OCR software works by recognizing pre-built characters the software was initially programmed to recognize. When a document does not follow that same pattern, the software cannot recognize it. Handwritten documents are a good example of this, in which the same letter may appear differently to the software, depending on how it was written. Some programs, such as ABBYY FineReader, have attempted to resolve this problem by incorporating a training program, which allows users to train the system to read specific types of handwriting. Even still, that training process requires human input, and there is still much work for individuals to put into ensuring that the processed document is accurate. As a result, OCR can be a time-consuming process that still requires plenty of human labor for a project.  

Western Biases  

Another key issue with the OCR process is the Western biases that went into the creation of the software. Many common OCR programs were designed to handle projects with Latinized scripts. While helpful for some projects, this left barriers to documents with non-Latinized scripts, particularly from languages commonly used outside the West. While advances have been made on this front, the advancements are still far behind that of Latinized scripts. For example, ABBYY FineReader is one of the few software programs that will scan in non-western languages, but it cannot incorporate its training program when those scripts aren’t Latinized. Adobe Acrobat can also scan documents with languages that use non-Latinized scripts, but its precision is less consistent than with those languages that do.  

An old version of ABBYY FineReader. The text scanned on the left is a language with a non-Latinized script. The right side shows a variety of errors due to the system's lack of knowledge of that language.
Photo Credit: Paul Tafford 

Addressing the Issues with OCR 

Although OCR has performed many amazing tasks, there is still much development needed when it comes to projects related to this aspect of scholarly research. One crucial component when considering taking on an OCR project is to recognize the limitations of the software and to account for that when determining the scope of your project. At this stage, OCR technology is certainly a time-saver and fundamentally changing the possibilities of scholarship, but without human input, these projects fail to make an impact. Likewise, recognizing the inequality of processing for non-western languages in some of the more prevalent OCR software (which several developers have looked to offset by creating OCR programs specifically catered to specific non-Latinized languages). Acknowledging these issues can help us consider the scope of various projects and also allow us to address these issues to make OCR a more accessible field.

Tech Teaching: Pedagogies for Teaching Technology and other Software Tools to Learners

Posted on by
Reply

When I began my role as a graduate assistant at the Scholarly Commons, my background in technology was extremely limited. As I have worked in this space, however, I have not only had the opportunity to learn how to use technology myself but teach others how to use these same tools through consultations and workshops. As technology begins to encompass more of our lives, I wanted to share a few tips for providing instruction focused on digital technology and software. While these pedagogies also apply to other teaching contexts, specific examples in this post will cater to digital technology.

Photo of two people working together on laptops.
Photo Credit: Christina Morillo

Active Learning

A crucial component of learning technology is allowing learners to directly engage with the technology they are looking to understand. By having direct engagement with the tool, learners will have a better grasp of how that tool works instead of just hearing about its functions in the abstract. If possible, it is highly encouraged that the instructional session has users access the technology or software they are learning, so that they can follow along as they experience how to navigate the tool. If that is not possible though, a few other alternatives may include watching the tool work from either the instructor conducting a live demonstration or finding a video directly showing the technology at work.

Scaffolding

Since technology is often complex, it is very easy for learners to feel overwhelmed by the sheer number of options and possibilities of what certain resources can do. Scaffolding as an instructional concept is a practice of designing a lesson that segments information into smaller sections that build upon each other. When providing instruction for a software program, for example, scaffolding may look like first helping users navigate the options of the tool, following that up with a basic function of the program, then performing a more complex task. Each of these steps is meant to build on one another and guide the learner by both showing them new aspects of the topic while incorporating previously acquired knowledge.

Photo of two people in front of a computer, with one older person guiding the hand of the younger person.
Photo Credit: August de Richelieu

Inclusive Learning

While inclusivity is valued in every learning environment, it is especially vital that instructors provide inclusive environments for teaching digital technology. Neglecting these principles will ultimately create barriers for certain users learning new technology. For general instruction sessions, applying universal design models will help streamline the process so that the session is accessible and meaningful for all types of learners. Considerations for font size when presenting to a workshop/classroom setting, for example, often help those with visual impairments follow along more easily, whereas not taking these considerations makes the learning process more difficult for them. Accommodating specific needs also helps to create an equitable environment that fosters learning for those whose needs may not be accounted for otherwise.

Using These Pedagogies in Personal Learning

Even if you are not planning on teaching others how to use technology, these same methods can also help you learn. Finding opportunities to engage with a particular tool hands-on will help you learn how to use it, rather than just reading articles abstractly about it. Likewise, breaking the content into smaller sections will help prevent overloading and help you progress in mastery of the tool. Finally, recognizing your needs as a learner and finding tools that are relevant to your needs will lift certain barriers to learning certain technologies. As you seek to learn and teach new technology, be creative and have fun with it!

Copyright Enforcement Tools as Censorship

Posted on by
Reply

This week, Scholarly Commons graduate assistants Zhaneille Green and Ryan Yoakum, alongside Copyright Librarian Sara Benson, appeared as guest writers for the International Federation of Library Associations and Institutions’ blog as part of a series for Copyright Week. Their blog post looks at how the current copyright tools on platforms such as YouTube and Facebook allow large corporate or governmental entities to silence and suppress individual voices. You can read the full blog post on the IFLA blog website.

A Non-Data Scientist’s Take on Orange

Posted on by
Reply

Introduction

Coming from a background in the humanities, I have recently developed an interest in data analysis but am just learning how to code. While I have been working to remedy that, one of my professors showed me this program known as Orange. Created in 1996, Orange is primarily designed to help researchers through the data analysis process, whether that is by applying machine learning methods or visualizing data. It is an open-source program (meaning you can download it for free!) and uses a graphical user interface (GUI) that allows the user to perform their analysis by matching icons to one another instead of having to write code.

How it Works

Orange works by using a series of icons known as widgets to perform the various functions that a user would otherwise need to manually code if they were using a program such as Python or R. Each widget appears as a bubble that can be moved around the interface. Widgets are divided into various categories based on the different steps in the analysis process. You can draw lines between the widgets to create a sequence, which will determine the process for how that data is analyzed (which is also known as a workflow). In its current state, Orange contains 96 widgets, each with different customizable and interactive components, so there are many opportunities for performing different types of basic data analysis with this software.

To demonstrate, I will use a dataset about the nutrition facts in specific foods (courtesy of Kaggle) to see how accurately a machine learner can predict the food group a given item falls in based on its nutrients. The following diagram is the workflow I designed to analyze this data:

This is the workflow I designed to analyze a sample sheet of data. From left to right, the widgets placed are "File," "Logistic Regression," "Test and Score," and "Confusion Matrix."

On the left side of the screen are different tabs that each contain a series of widgets related to the task at hand. By clicking on the specific widgets, a pop-up window appears that allows you to interact with the widget. In this particular workflow, the “file” widget is where I can upload the file I want to analyze (there are a lot of different formats you can upload too; in this case, I uploaded an Excel spreadsheet). From there, I chose the machine learning method that I wanted to use to classify the data. The third widget tests the data using the classification method, and compares it to the original data. Finally, the results are visualized through the “confusion matrix” widget to show which cases the machine learner accurately predicted and which ones it got wrong.

A confusion matrix of the predicted classification of food items based on the amount of nutrients in them compared to the actual classifications .

The Limitations

While Orange is a helpful tool for those without a coding background, this system also presents some limitations when it comes to performing certain types of data analysis. One way Orange tries to reconcile this is by providing a widget where the user can insert some Python script into the workflow. While this feature may be helpful for those with a coding background, it would not really impact those who do not have a coding background, thereby limiting the ways they can analyze data.

Additionally, although Orange can visualize data, there are not many features that allow users to adjust the visualization’s appearance. Such limitations may require exporting the data and using another tool to create a more accessible or visually appealing data visualization, but for now, Orange is quite limited in this capacity. As a result, Orange is an incredibly useful tool for basic data visualization but struggles with more advanced types of data science work that may require using other tools or programming to accomplish.

Final Remarks

If you are looking to get involved in data analysis but are just starting to develop an interest in coding, then Orange is a great tool to use. Unlike most data analysis programs, the user-designed interface of Orange makes it easy to perform basic types of data analysis through its widgets. It is far from perfect though, and a lack of a coding background is going to limit the ways you can analyze and visualize your data. Nevertheless, Orange can be an incredibly useful tool if you are just starting to learn how to code and looking to understand the basics of data science!

Welcome Back to the Scholarly Commons!

Posted on by
Reply

The Scholarly Commons is excited to announce we have merged with the Media Commons! Our units have united to provide equitable access to innovative spaces, digital tools, and assistance for media creation, data visualization, and digital storytelling. We launched a new website this summer, and we’re thrilled to announce a new showcase initiative that highlights digital projects created by faculty and students. Please consider submitting your work to be featured on our website or digital displays. 

Looking to change up your office hours? Room 220 in the Main Library is a mixed-used space with comfortable seating and access to computers and screen-sharing technology that can be a great spot for holding office hours with students. 

Media Spaces

We are excited to announce new media spaces! These spaces are designed for video and audio recordings and equipped to meet different needs depending on the type of production. For quick and simple video projects, Room 220 has a green-screen wall on the southeast side of the room (adjacent to the Reading Room). The space allows anyone to have fun with video editing. You can use your phone to shoot a video of yourself in front of the green wall and use software to replace the green with a background of your choosing to be transported anywhere. No reservations required.

Green Screen Wall in Room 220. Next to it is some insignificant text for design purposes.

For a sound-isolated media experience, we are also introducing Self-Use Media Studios in Rooms 220 and 306 of the Main Library. These booths will be reservable and are equipped with an M1 Mac Studio computer, two professional microphones, 4K video capture, dual color-corrected monitors, an additional large TV display, and studio-quality speakers. Record a podcast or voiceover, collect interviews or oral histories, capture a video or give a remote stream presentation, and more at the Self-Use Media Studios.

Finally, we are introducing the Video Production Studio in Room 308. This is a high-end media creation studio complete with two 6K cameras, an 4K overhead camera, video inputs for computer-based presentation, professional microphones, studio-lighting, multiple backdrops, and a live-switching video controller for real-time presentation capture or streaming. Additionally, an M1 Mac Studio computer provides plenty of power to enable high-resolution video project editing. The Video Production Studio can be scheduled by arranged appointment and will be operated by Scholarly Commons staff once the space is ready to open. 

Stay tuned to our spaces page for more information about reserving these resources.

Loanable Tech

The Scholarly and Media Commons are pleased to announce the re-opening of loanable technology in Room 306 of the Main Library. Members of the UIUC community can borrow items such as cameras, phone chargers, laptops, and more from our loanable technology desk. The loanable technology desk is open 10:30 a.m. – 7:30 p.m. Mondays-Thursdays, 10:30 a.m. – 5:30 p.m. Fridays, and 2-6:30 p.m. on Sundays. Check out the complete list of loanable items for more on the range of technology we provide.

Drop-in Consultation Hours

Drop-in consultations have returned to Room 220. Consultations this semester include:

  • GIS with Wenjie Wang – Tuesdays 1 – 3 p.m. in Consultation Room A.
  • Copyright with Sara Benson – Tuesdays 11 a.m. – 12 p.m. in Consultation Room A.
  • Media and design with JP Goguen – Thursdays 10 a.m. – 12 p.m. in Consultation Room A.
  • Data analysis with the Cline Center for Advanced Social Research – Thursdays 1 – 3 p.m. in Consultation Room A.
  • Statistical consulting with the Center for Innovation, Technology, and Learning (CITL) – 10 a.m. – 5 p.m. Mondays, Tuesdays, Thursdays, and Fridays, as well as 10 a.m. – 4 p.m. Wednesdays in Consultation Room B.

Finally, a Technology Services help desk has moved into Room 220. They are available 10 a.m. – 5 p.m. Mondays-Fridays to assist patrons with questions about password security, email access, and other technology needs.

Spatial Computing and Immersive Media Studio

Later this fall, we will launch the Spatial Computing and Immersive Media Studio (SCIM Studio) in Grainger Library. SCIM Studio is a black-box space focused on emerging technologies in multimedia and human-centered computing. Equipped with 8K 360 cameras, VR and AR hardware, a 22-channel speaker system, Azure Kinect Depth Cameras, Greenscreen, and a Multi-Camera and display system for Video Capture & Livestreaming, SCIM Studio will cater to researchers and students interested in utilizing the cutting edge of multimedia technology. The Core i9 workstation equipped with Nvidia A6000 48GB GPU will allow for 3D modeling, Computer Vision processing, Virtual Production compositing, Data Visualization/Sonification, and Machine Learning workflows. Please reach out to Jake Metz if you have questions or a project you would like to pursue at the SCIM Studio and keep your eye on our website for launch information. 

Have Questions?

Please continue to contact us through email (sc@library.illinois.edu) for any questions about the Scholarly and Media Commons this year. Finally, you can check out the new Scholarly Commons webpage for more information about our services, as well as our staff directory to set up consultations for specific services. 

We wish you all a wonderful semester and look forward to seeing you here at the Scholarly and Media Commons!