Matching Items (125)
Filtering by
- All Subjects: Computer Science
- All Subjects: Creative Project
- Creators: Baral, Chitta
- Creators: Walter Cronkite School of Journalism and Mass Communication
Description
Virtual digital assistants are automated software systems which assist humans by understanding natural languages such as English, either in voice or textual form. In recent times, a lot of digital applications have shifted towards providing a user experience using natural language interface. The change is brought up by the degree of ease with which the virtual digital assistants such as Google Assistant and Amazon Alexa can be integrated into your application. These assistants make use of a Natural Language Understanding (NLU) system which acts as an interface to translate unstructured natural language data into a structured form. Such an NLU system uses an intent finding algorithm which gives a high-level idea or meaning of a user query, termed as intent classification. The intent classification step identifies the action(s) that a user wants the assistant to perform. The intent classification step is followed by an entity recognition step in which the entities in the utterance are identified on which the intended action is performed. This step can be viewed as a sequence labeling task which maps an input word sequence into a corresponding sequence of slot labels. This step is also termed as slot filling.
In this thesis, we improve the intent classification and slot filling in the virtual voice agents by automatic data augmentation. Spoken Language Understanding systems face the issue of data sparsity. The reason behind this is that it is hard for a human-created training sample to represent all the patterns in the language. Due to the lack of relevant data, deep learning methods are unable to generalize the Spoken Language Understanding model. This thesis expounds a way to overcome the issue of data sparsity in deep learning approaches on Spoken Language Understanding tasks. Here we have described the limitations in the current intent classifiers and how the proposed algorithm uses existing knowledge bases to overcome those limitations. The method helps in creating a more robust intent classifier and slot filling system.
In this thesis, we improve the intent classification and slot filling in the virtual voice agents by automatic data augmentation. Spoken Language Understanding systems face the issue of data sparsity. The reason behind this is that it is hard for a human-created training sample to represent all the patterns in the language. Due to the lack of relevant data, deep learning methods are unable to generalize the Spoken Language Understanding model. This thesis expounds a way to overcome the issue of data sparsity in deep learning approaches on Spoken Language Understanding tasks. Here we have described the limitations in the current intent classifiers and how the proposed algorithm uses existing knowledge bases to overcome those limitations. The method helps in creating a more robust intent classifier and slot filling system.
ContributorsGarg, Prashant (Author) / Baral, Chitta (Thesis advisor) / Kumar, Hemanth (Committee member) / Yang, Yezhou (Committee member) / Arizona State University (Publisher)
Created2018
Description
The purpose of this creative project is to make an E-Book that promotes time management for college students in a way that interests them. The author of this recognizes that there are many distractions to keep college students from sitting down and reading a textbook; that is why an E-Book featuring videos and interactive videos was chosen. The research questions presented below began my research and understanding of the topic. These questions are as follows: 1. What is a way to promote time management for college students? a) What are some mediums that will appeal to young people who want to do more than just read a book. 2. When figuring out how to manage their time, what are the areas of life students consider to be most important? 3. What perspectives to various facets of the world like, business, academia and the foreign community think about time management? 4. What perspective to millennials have on time management? By answering these questions above, the author hopes to understand what is good time management, and how to explore it in a way that will interest young people. The author is doing so by creating a series of narrative videos that he himself acted in portraying a fictitious student both engaging in and not practicing good time management techniques. The created nine videos, with three dedicated to a section each. The three sections were what students do wrong, how they can improve and how they can maintain their success. Within each section were three sub- sections that students must use time management skills for: mental techniques, physical well-being, and juggling work and personal commitments. See the attached documents (Appendix A) for a full collection of the scripts that were created for these videos. The author also created quizzes through the website Bookry, allowing him to make review questions for those reading the book. The quizzes were then made into widgets and inserted into the book. Each quiz was about 5 questions each and was at the end of each of the sub-sections, meaning there were 45 questions total. See the attached documents (Appendix B) for screenshots of each quiz question and the correct answer.
ContributorsCzajka, Jagger James (Author) / Silcock, Bill (Thesis director) / Rodriguez, Rick (Committee member) / Dodge, Nancie (Committee member) / Walter Cronkite School of Journalism and Mass Communication (Contributor) / Barrett, The Honors College (Contributor)
Created2016-05
Description
This thesis analyzes identity construction through street style fashion in the city. The focus of this project is Roosevelt Row, the artists' district in Downtown Phoenix. The goal of this project is to compare Roosevelt Row's marketing image with the fashion seen on the streets and at events in the area. The creative project involved the creation of an iPad publication displaying the street style fashions seen on Roosevelt Row. This project aims to analyze if the street style fashion seen on Roosevelt Row reflects the marketing image of the area.
ContributorsKraus, Tamara Renee (Author) / Barrett, Marianne (Thesis director) / Gilpin, Dawn (Committee member) / Barrett, The Honors College (Contributor) / Department of Marketing (Contributor) / Walter Cronkite School of Journalism and Mass Communication (Contributor)
Created2015-12
Description
"The Process and Analysis of Film Reviewing" is a two-part project that discovers what goes into the film reviewing process and what produces a good film review. A film review website entitled "Reel Reviews" (reelfilmreviews.com) was created, which features original content such as film reviews, trailers and information about special features and events relevant to the film industry. The paper portion of the project takes an analytical look at the current state of the film reviewing world and addresses the changes in technology that have an impact on how film reviewing is done.
ContributorsRepasi, Bianca Joyce (Author) / Baker, Aaron (Thesis director) / Dodge, Nancie (Committee member) / Barrett, The Honors College (Contributor) / Walter Cronkite School of Journalism and Mass Communication (Contributor)
Created2015-05
Description
The purpose of this project is to use powerful visual storytelling techniques to convey a social need and an effective solution. Guatemala is a third world country, where poverty is widespread and the birth rate is high. Among the most economically and educationally disadvantaged are the Mayan women. Arizona nun, Sister Marife Hellman, recognized the needs of this population and founded a school to serve them. Hellman's mission is to provide a quality education to those underserved, so they can become positive leaders in their native communities. The website and video materials produced for this thesis are meant to be used for fundraising purposes on behalf of the school. All funds raised will help Hellman's alumni launch schools in their native areas, giving access to education that has long been nonexistent. Watch the mini-documentary here: www.youtube.com/watch?v=LxTiuwQCH44&t=17s.
ContributorsLaduke, Sierra Dawn (Author) / Craft, John (Thesis director) / Mork, Nick (Committee member) / Walter Cronkite School of Journalism and Mass Communication (Contributor) / School of Community Resources and Development (Contributor) / Barrett, The Honors College (Contributor)
Created2018-05