17 research outputs found
Automatically extracting interaction and app data from mobile application traces
No full textIn this research, we used an existing system to collect mobile interaction traces and extract meaningful information in terms of interaction data, apps, and layout information and complexity of mobile apps. The preeminent driving force for this research was to come up with a system that is scalable and can be used to extract interactions and layouts from mobile apps, as well as enable us to make claims about the complexity of mobile apps and the flows that they offer. Throughout the course of this research, we collected Android mobile interaction traces and presented a technique which enables extraction of frequent interactive elements from the traces in an unsupervised manner using neural network auto-encoders and k-means clustering. The research work also enables us to find similar layouts across apps and make claims about the location of some of these interactive elements. This research provides a scalable data-driven approach to finding clusters of frequent icons and interactions as well as layouts.Submission published under a 24 month embargo labeled 'U of I Access', the embargo will last until 2018-05-01The student, Abhishek Harish, accepted the attached license on 2016-04-14 at 15:22.The student, Abhishek Harish, submitted this Thesis for approval on 2016-04-14 at 15:27.This Thesis was approved for publication on 2016-04-15 at 13:54.DSpace SAF Submission Ingestion Package generated from Vireo submission #9229 on 2016-07-07 at 13:49:16Made available in DSpace on 2016-07-07T20:27:21Z (GMT). No. of bitstreams: 2
HARISH-THESIS-2016.pdf: 15467630 bytes, checksum: f215401b9a562e8e4fcb47aa44a10efa (MD5)
LICENSE.txt: 4212 bytes, checksum: 63cdef903777511bbade5d70c0a9322d (MD5)
Previous issue date: 2016-04-15Embargo set by: Seth Robbins for item 93109
Lift date: 2018-07-07T20:28:14Z
Reason: Author requested U of Illinois access only (OA after 2yrs) in Vireo ETD systemEmbargo set by: Seth Robbins for item 93109
Lift date: 2018-07-07T20:35:34Z
Reason: Author requested U of Illinois access only (OA after 2yrs) in Vireo ETD systemU of I Only Restriction Lifted for Item 93109 on 2018-07-08T09:15:09Z
Generative models for predictive UI design tools
No full textUser interface (UI) design is a central part of the mobile app creation process, which involves specifying the elements that should be placed on a screen, and how they should be arranged and styled. This paper introduces a generative model approach to predictive design for mobile UI layouts. Given a partial UI design, the model predicts the next UI element that should be added to the layout. Moreover, the model can be used queried multiple times in succession to autocomplete an entire UI screen. To power this design interaction, we present two types of models: generative adversarial networks (GANs) [7] and variational auto-encoders (VAEs) [15]. We train the GAN and VAE models over 1949 mobile UIs that represent a variety of screen types (e.g. Login, Onboarding), and compare both models along standard and design-based metrics, identifying key tradeoffs. Finally, we present a mobile UI mockup tool that leverages the GAN-based model to support a predictive design workflow.Submission published under a 24 month embargo labeled 'U of I Access', the embargo will last until 2021-05-01The student, Jason Situ, accepted the attached license on 2019-04-25 at 21:57.The student, Jason Situ, submitted this Thesis for approval on 2019-04-25 at 22:10.This Thesis was approved for publication on 2019-04-26 at 11:13.DSpace SAF Submission Ingestion Package generated from Vireo submission #13938 on 2019-08-22 at 15:08:48Made available in DSpace on 2019-08-23T20:36:13Z (GMT). No. of bitstreams: 2
SITU-THESIS-2019.pdf: 2953945 bytes, checksum: 3d17819257a18203fc430721072ede49 (MD5)
LICENSE.txt: 4207 bytes, checksum: 192b15e610c8807c1dc9eff3eff5b94b (MD5)
Previous issue date: 2019-04-26Embargo set by: Seth Robbins for item 112226
Lift date: 2021-08-23T20:36:18Z
Reason: Author requested U of Illinois access only (OA after 2yrs) in Vireo ETD systemU of I Only Restriction Lifted for Item 112226 on 2021-08-24T09:15:24Z
Curation and privacy in mobile application UI repositories
No full textMobile interaction mining allows everyday interaction data to be mined for insights into the best performing design patterns, usability problems, and overall design trends. So far, this data has primarily come from automated application exploration or crowdworkers completing smartphone tasks as part of a study. Both of these methods have a primary issue that the interaction patterns do not quite align with how everyday users interact with applications. However, mining interaction traces that contain personally identifiable information from real users presents a problem, mainly when that data is to be published. This thesis provides an exploratory look at the data curation and privacy considerations required to share mobile application interaction data publicly. Regarding the privacy side, we will focus on applications in the Finance and Health categories.Submission published under a 24 month embargo labeled 'U of I Access', the embargo will last until 2021-08-01The student, Oliver Melvin, accepted the attached license on 2019-07-19 at 00:13.The student, Oliver Melvin, submitted this Thesis for approval on 2019-07-19 at 00:16.This Thesis was approved for publication on 2019-07-19 at 08:23.DSpace SAF Submission Ingestion Package generated from Vireo submission #14387 on 2019-11-26 at 13:06:22Made available in DSpace on 2019-11-26T20:49:36Z (GMT). No. of bitstreams: 2
MELVIN-THESIS-2019.pdf: 14089411 bytes, checksum: 920e1280723f975b4246baa8eb052e33 (MD5)
LICENSE.txt: 4210 bytes, checksum: ed8a492b7b20ee6c4045e3762ec240c4 (MD5)
Previous issue date: 2019-07-19Embargo set by: Seth Robbins for item 112987
Lift date: 2021-11-26T20:49:41Z
Reason: Author requested U of Illinois access only (OA after 2yrs) in Vireo ETD systemU of I Only Restriction Lifted for Item 112987 on 2021-11-27T10:15:30Z
Mobile design semantics
No full textGiven the growing number of mobile apps and their increasing impact on modern life, researchers have developed black-box approaches to mine mobile app design and interaction data. Although the data captured during interaction mining is descriptive, it does not expose the design semantics of UIs: what elements on the screen mean and how they are used. This thesis introduces an automatic approach for semantically annotating the elements comprising a UI given the data captured during interaction mining. Through an iterative open coding of 73k UI elements and 720 screens, we first created a lexical database of 24 types of UI components, 197 text button concepts, and 135 icon classes shared across apps. Using the labeled data created during this process, we learned code-based patterns to detect components, and trained a convolutional neural network which distinguishes between 99 icon classes with 94% accuracy. With this automated approach, we computed semantic annotations for the 72k unique UIs comprising the Rico dataset, assigning labels for 78% of the total visible, non-redundant elements.Submission published under a 24 month embargo labeled 'U of I Access', the embargo will last until 2020-05-01The student, Thomas Liu, accepted the attached license on 2018-04-27 at 16:08.The student, Thomas Liu, submitted this Thesis for approval on 2018-04-27 at 16:13.This Thesis was approved for publication on 2018-04-27 at 16:36.DSpace SAF Submission Ingestion Package generated from Vireo submission #12547 on 2018-08-31 at 17:21:52Made available in DSpace on 2018-09-04T20:42:02Z (GMT). No. of bitstreams: 2
LIU-THESIS-2018.pdf: 14557319 bytes, checksum: 53aab5cbc6fed062e55ef34562a4fa27 (MD5)
LICENSE.txt: 4207 bytes, checksum: f25209dcc54c5ef7100df8d0321824ec (MD5)
Previous issue date: 2018-04-27Embargo set by: Seth Robbins for item 107331
Lift date: 2020-09-04T20:42:08Z
Reason: Author requested U of Illinois access only (OA after 2yrs) in Vireo ETD systemU of I Only Restriction Lifted for Item 107331 on 2020-09-05T09:15:23Z
A large-scale study of fashion influencers on Twitter
No full textThe rise of social media has changed the nature of the fashion industry. Influence is no longer concentrated in the hands of an elite few: social networks distribute power across a broad set of tastemakers; trends are driven bottom-up and top-down; and designers, retailers, and consumers are regularly inundated with new styles and looks.
This thesis presents a large-scale study of fashion influencers on Twitter and proposes a fashion graph visualization dashboard to explore the social interactions between these Twitter accounts. Leveraging a dataset of 11.5k Twitter fashion accounts, a content-based classifier was trained to predict which accounts are fashion-centric. With the classifier, I identified more than 300k fashion-related accounts through a snowball crawling and then defined a stable group of 1000 influencers as the fashion core. I further human-labeled these influencers’ Twitter accounts and mine their recent tweets. Finally, I built a fashion graph visualization dashboard that allows users to visualize the interactions and relationships between individuals, brands, and media influencers.Submission published under a 24 month embargo labeled 'U of I Access', the embargo will last until 2021-05-01The student, Qinglin Chen, accepted the attached license on 2019-04-25 at 15:13.The student, Qinglin Chen, submitted this Thesis for approval on 2019-04-25 at 15:19.This Thesis was approved for publication on 2019-04-26 at 14:13.DSpace SAF Submission Ingestion Package generated from Vireo submission #13922 on 2019-08-22 at 15:08:43Made available in DSpace on 2019-08-23T20:36:12Z (GMT). No. of bitstreams: 2
CHEN-THESIS-2019.pdf: 10256513 bytes, checksum: c4ff0b8f2db78d40d23868675deca11b (MD5)
LICENSE.txt: 4209 bytes, checksum: d887e9034c676c149aa0fbfc0932c6f5 (MD5)
Previous issue date: 2019-04-26Embargo set by: Seth Robbins for item 112221
Lift date: 2021-08-23T20:36:18Z
Reason: Author requested U of Illinois access only (OA after 2yrs) in Vireo ETD systemU of I Only Restriction Lifted for Item 112221 on 2021-08-24T09:15:34Z
Aqueduct: Task-based entry points in Android apps
No full textModern smartphones offer voice assistants to ease a variety of tasks. However, the actions that can be performed by current voice assistants are limited – a predefined set of built in actions like checking the weather, and a few hooks that can be built into third-party applications. To extend assistant actions to third-party applications, the onus is on the application developers to manually add support for voice assistant integration. To improve the link between voice assistants and third-party apps, we built Aqueduct, a data driven task-based app search and task entry point discovery system for Android. We search over app UI data augmented with semantic annotations to find applications and screens within those applications that can accomplish a given task. Furthermore, Aqueduct can leverage the package name and the activity name of the discovered screen to automatically navigate users to that screen. A user study was conducted to compile a set of common smartphone tasks and evaluate the effectiveness of Aqueduct, which showed that it is effective at finding task-based entry points for a wide range of tasks.
Aqueduct is also useful for augmenting search in application repositories, finding starting points for execution for task-automation systems, and even generating deep link suggestions for applications.Submission published under a 24 month embargo labeled 'U of I Access', the embargo will last until 2021-08-01The student, Aravind Sagar, accepted the attached license on 2019-07-15 at 12:17.The student, Aravind Sagar, submitted this Thesis for approval on 2019-07-15 at 12:24.This Thesis was approved for publication on 2019-07-15 at 14:11.DSpace SAF Submission Ingestion Package generated from Vireo submission #14321 on 2019-11-26 at 13:05:54Made available in DSpace on 2019-11-26T20:49:30Z (GMT). No. of bitstreams: 2
SAGAR-THESIS-2019.pdf: 8647624 bytes, checksum: f94f0b48d3da9aea55f3691b1b53ec8d (MD5)
LICENSE.txt: 4210 bytes, checksum: ab2ffd65bf700f93649c6a0e21d7c63a (MD5)
Previous issue date: 2019-07-15Embargo set by: Seth Robbins for item 112966
Lift date: 2021-11-26T20:49:41Z
Reason: Author requested U of Illinois access only (OA after 2yrs) in Vireo ETD systemU of I Only Restriction Lifted for Item 112966 on 2021-11-27T10:15:27Z
Enabling type differentiated explainable queries across modalities for different fashion items
No full textOne of the biggest differences between shopping online and in person is the limited scope and expressibility of the queries that current systems allow and can handle. In person, users often employ a combination of linguistic and visual tools at their disposal to create complex queries. Handling such queries requires modeling relationships between products of the same type, products of different types, products and outfits, and products and their attributes. In this paper, we propose a system that models these relationships by: (i) building a robust visual representation of items that captures notions of similarity and compatibility between products, (ii) learning to predict low-level (color, type) and high-level (style, brand) attributes of the items from their visual representations, and (iii) learning segment-wise maps of outfits to items. For each part, we evaluate the model by demonstrating its performance on relevant tasks like outfit completion, item retrieval, etc., and flexibility through example results for complex queries.Submission published under a 24 month embargo labeled 'U of I Access', the embargo will last until 2021-05-01The student, Krishna Dusad, accepted the attached license on 2019-04-26 at 11:30.The student, Krishna Dusad, submitted this Thesis for approval on 2019-04-26 at 11:39.This Thesis was approved for publication on 2019-04-26 at 15:04.DSpace SAF Submission Ingestion Package generated from Vireo submission #13945 on 2019-08-22 at 15:08:49Made available in DSpace on 2019-08-23T20:36:13Z (GMT). No. of bitstreams: 2
DUSAD-THESIS-2019.pdf: 17414255 bytes, checksum: 2f966e1c46cf63a5e1f818854856ca87 (MD5)
LICENSE.txt: 4210 bytes, checksum: 35e79fa00d134c01166c88270764c0f8 (MD5)
Previous issue date: 2019-04-26Embargo set by: Seth Robbins for item 112227
Lift date: 2021-08-23T20:36:18Z
Reason: Author requested U of Illinois access only (OA after 2yrs) in Vireo ETD systemU of I Only Restriction Lifted for Item 112227 on 2021-08-24T09:15:38Z
A Novel and Efficient De-duplication System for HDFS
No full textAbstractBig Data is a frequent generation and updating of large volume of data around the clock across the globe by the users. Handling large volume of data in a real time environment is a challenging task. Distributed File System is one of the strategy to handle large volume of data in the real time. Distributed file system is a collection of independent computers that appear to the users of the system as a single coherent system. In Distributed file system common files can be shared between the nodes, the drawbacks are scalability, replication, availability and very expensive to buy a hardware server. To overcome this issue Hadoop Distributed File System came into existence. Hadoop distributed file system to run on cluster of commodity hardware like personal computer and laptop. HDFS provides the scalable, fault-tolerance, cost-efficient storage for Bigdata. Hadoop Distributed File System support data duplication to achieve high data reliability. However additional utilization of storage space is required due to duplication strategy. HDFS Storage space can be managed efficiently by implementing De-duplication techniques. The objective of the research is to eliminate file duplication by implementing De-duplication strategy. A novel and efficient De-duplication system using HDFS approach is introduced in this research work. To implement De-duplication strategy, hash values are computed for files using MD5 and SHA1 algorithms . The generated hash value for a file is checked with the existing file to identify the presence of duplication. If duplication exists, the system will not allow the user to upload the duplicate copy to the HDFS. Hence memory utilization is handled efficiently in HDFS
Change in shape of the crystallite size with wood flour and their native cellulose using WAXS studies
No full textEnzymatic Liquefaction of Jackfruit (Artocarpus Heterophyllus Lam.) Pulp for Juice Production and value addition thereafter
No full textThis Dissertation / Report is the outcome of investigation carried out by the creator(s) / author(s) at the department/division of Central Food Technological Research Institute (CFTRI), Mysore mentioned below in this page
