Today there is no effective support for device-wide question answer- ing on mobile devices. State-of-the-art QA models are deep learning behemoths designed for the cloud which run extremely slow and require more memory than available on phones. We present DeQA, a suite of latency- and memory- optimizations that adapts existing QA systems to run completely locally on mobile phones. Specifi- cally, we design two latency optimizations that (1) stops processing documents if further processing cannot improve answer quality, and (2) identifies computation that does not depend on the ques- tion and moves it offline. These optimizations do not depend on the QA model internals and can be applied to several existing QA models. DeQA also implements a set of memory optimizations by (i) loading partial indexes in memory, (ii) working with smaller units of data, and (iii) replacing in-memory lookups with a key-value database. We use DeQA to port three state-of-the-art QA systems to the mobile device and evaluate over three datasets. The first is a large scale SQuAD dataset defined over Wikipedia collection. We also create two on-device QA datasets, one over a publicly available email data collection and the other using a cross-app data collection we obtain from two users. Our evaluations show that DeQA can run QA models with only a few hundred MBs of memory and provides at least 13x speedup on average on the mobile phone across all three datasets.
more »
« less
Mining, analyzing, and modeling text written on mobile devices
Abstract We present a method for mining the web for text entered on mobile devices. Using searching, crawling, and parsing techniques, we locate text that can be reliably identified as originating from 300 mobile devices. This includes 341,000 sentences written on iPhones alone. Our data enables a richer understanding of how users type “in the wild” on their mobile devices. We compare text and error characteristics of different device types, such as touchscreen phones, phones with physical keyboards, and tablet computers. Using our mined data, we train language models and evaluate these models on mobile test data. A mixture model trained on our mined data, Twitter, blog, and forum data predicts mobile text better than baseline models. Using phone and smartwatch typing data from 135 users, we demonstrate our models improve the recognition accuracy and word predictions of a state-of-the-art touchscreen virtual keyboard decoder. Finally, we make our language models and mined dataset available to other researchers.
more »
« less
- Award ID(s):
- 1750193
- NSF-PAR ID:
- 10157980
- Date Published:
- Journal Name:
- Natural Language Engineering
- ISSN:
- 1351-3249
- Page Range / eLocation ID:
- 1 to 33
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
null (Ed.)Typing every character in a text message may require more time or effort than strictly necessary. Skipping spaces or other characters may be able to speed input and also reduce a user's physical input effort. This can be particularly important for people with motor impairments. In a large crowdsourced study, we found workers frequently abbreviated text by omitting mid-word vowels. We designed a recognizer optimized for noisy input where users often omit spaces and mid-word vowels. We show using neural language models for selecting training text and rescoring sentences improved accuracy. On noisy touchscreen data collected from hundreds of users, we found accurate abbreviated input was possible even if a third of characters were omitted. Finally, in a study where users had to dwell for a second on each key, sentence abbreviated input was competitive with a conventional keyboard with word predictions. After practice, users wrote abbreviated sentences at 9.6 words-per-minute versus word input at 9.9 words-per-minute.more » « less
-
null (Ed.)The prevalence of mobile phones and wearable devices enables the passive capturing and modeling of human behavior at an unprecedented resolution and scale. Past research has demonstrated the capability of mobile sensing to model aspects of physical health, mental health, education, and work performance, etc. However, most of the algorithms and models proposed in previous work follow a one-size-fits-all (i.e., population modeling) approach that looks for common behaviors amongst all users, disregarding the fact that individuals can behave very differently, resulting in reduced model performance. Further, black-box models are often used that do not allow for interpretability and human behavior understanding. We present a new method to address the problems of personalized behavior classification and interpretability, and apply it to depression detection among college students. Inspired by the idea of collaborative-filtering, our method is a type of memory-based learning algorithm. It leverages the relevance of mobile-sensed behavior features among individuals to calculate personalized relevance weights, which are used to impute missing data and select features according to a specific modeling goal (e.g., whether the student has depressive symptoms) in different time epochs, i.e., times of the day and days of the week. It then compiles features from epochs using majority voting to obtain the final prediction. We apply our algorithm on a depression detection dataset collected from first-year college students with low data-missing rates and show that our method outperforms the state-of-the-art machine learning model by 5.1% in accuracy and 5.5% in F1 score. We further verify the pipeline-level generalizability of our approach by achieving similar results on a second dataset, with an average improvement of 3.4% across performance metrics. Beyond achieving better classification performance, our novel approach is further able to generate personalized interpretations of the models for each individual. These interpretations are supported by existing depression-related literature and can potentially inspire automated and personalized depression intervention design in the futuremore » « less
-
Text entry is a common and important part of many intelligent user interfaces. However, inferring a user’s intended text from their input can be challenging: motor actions can be imprecise, input sensors can be noisy, and situations or disabilities can hamper a user’s perception of interface feedback. Numerous prior studies have explored input on touchscreen phones, smartwatches, in midair, and on desktop keyboards. Based on these prior studies, we are releasing a large and diverse data set of noisy typing input consisting of thousands of sentences written by hundreds of users on QWERTY-layout keyboards. This paper describes the various subsets contained in this new research dataset as well as the data format.more » « less
-
Mobile computing devices are widely used in our daily life. With their increased use, a large amount of sensitive data are collected, stored, and managed in the mobile devices. To protect sensitive data, encryption is often used but, traditional encryption is vulnerable to coercive attacks in which the device owner is coerced by the adversary to disclose the decryption key. To defend against the coercive attacks, Plausibly Deniable Encryption (PDE) has been designed which can allow the victim user to deny the existence of hidden sensitive data. The PDE systems have been explored broadly for smartphones. However, the PDE systems which are suitable for wearable mobile devices are still missing in the literature. In this work, we design MobiWear, the first PDE system specifically for wearable mobile devices. To accommodate the hardware nature of wearable devices, MobiWear: 1) uses image steganography to achieve PDE, which suits the resource-limited wearable devices; and 2) relies on various sensors equipped with the wearable devices to input passwords, rather than requiring users to enter them via a keyboard or a touchscreen. Security analysis and experimental evaluation using a real-world prototype (ported to an LG G smartwatch) show that MobiWear can ensure deniability with a small computational overhead as well as a small decrease of image quality.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- https://doi.org/10.1017/S1351324919000548
