Browsing by Author "Barr, Christopher D."
Now showing 1 - 5 of 5
- Results Per Page
- Sort Options
Item Automatic Keyword Detection for Text Summarization(2019-05) Koka, Raga Shalini 1994-; Subhlok, Jaspal; Solorio, Thamar; Barr, Christopher D.Lecture videos are extremely useful and great learning companions for students. The ICS (Indexed, Captioned, and Searchable) video project provides students a flexible way to navigate across the lectures by automatically dividing the lecture into topical segments. Presenting keywords to every segment can provide an overview of the content discussed in a segment and improve navigation. Identifying keywords manually requires human effort and consumes a lot of time for lecture videos that are typically an hour or longer. This thesis proposes methods to automatically detect keywords to summarize the content in a video segment. The input to the keyword detection algorithm is text from the video frames extracted by OCR, and I enhance the text with auto-correction in a post-processing pass. Automatically detecting keywords is challenging as the importance of a word depends on a variety of factors such as frequency, font size, and duration of time it is present on the screen. Other factors include relative frequency in a video segment versus the rest of the video and domain significance derived from external sources. This thesis explores how these factors contribute to the importance of a word and how they can be combined to identify good keywords. I evaluated the performance of the proposed methods by comparing the keywords generated by the algorithm with the tags chosen by experts on 121 segments of 11 videos from different departments like Computer Science, Biology, and Biochemistry. I initialized the features to different combinations of weights and computed metrics like precision, recall, F1, BLEU score, and correlation scores. I also presented an analysis of errors and different areas that can be explored to generate higher quality keywords.Item Diagnostic and Construct Validation of Symptom and Performance Validity Tests of Malingering in a Civil Litigation Context(2018-08) Fox, Katherine Alvord; Vincent, John P.; Harris, Gerald E.; Barr, Christopher D.The extent to which persons may malinger psychiatric symptoms is a legitimate concern in civil litigation. The consequences inherent in personal injury cases involving psychological distress necessitate an understanding of how malingering presents in medico-legal contexts and the validity and usefulness of available methods to detect malingering. The present study evaluated the construct and diagnostic validity of symptom-based (SVT) and performance-based (PVT) measures of malingering in a simulated personal injury paradigm. We evaluated the interrelationships between malingering measures and whether these measures were able to discriminate between “honest responders” and “malingerers.” Using a simulation design, 411 undergraduates were randomly assigned into four experimental conditions, which outlined the experience of a motor vehicle accident and subsequent psychological and cognitive symptoms. Conditions varied on the degree of suggestion to malinger symptoms as related to a personal injury case. Under this paradigm, participants completed measures of malingered symptomatology, including the TOMM, M-FAST, SIMS, and TSI-2 ATR. As predicted, we found weaker correlations between PVT and SVTs, but moderate significant correlations were found across symptom validity measures. These findings support conceptualization of malingering as a non-unitary construct. Results from ROC analysis suggest that only the TSI-2 ATR was useful in discriminating between simulation groups. Contrary to expectations, prominent measures of malingering (TOMM, M-FAST, and SIMS) did not discriminate between groups. Results may mean that these tests may operate differently than intended within a civil litigation context and depending on the type of psychopathology feigned.Item Evaluation of Speech And Text-Based Indexing For Classroom Lecture Videos(2014-12) Joshi, Mahima 1990-; Subhlok, Jaspal; Johnson, Olin; Barr, Christopher D.Lecture videos are useful and great learning resources. At the University of Houston, videos are widely used throughout departments within the College of Natural Sciences and Mathematics such as Computer Science, Biology and Biochemistry, Earth and Atmospheric Sciences, etc. Since most videos are very long, it is difficult to directly access the required topic within a video. The ICS (indexed, captioned, and searchable) videos project provides students direct access to a topic within video lectures by providing index points representing the topic. These index points are generated using text from the extracted images using OCR (optical character recognition) technology. Index points are assigned with the assistance of an indexing algorithm that determines topic change based on text similarity. We present a topic-based lecture video segmentation using speech text/captions. The purpose of this thesis is to utilize the spoken text of a lecture video to assign index points using an underlying text-based indexing algorithm. To achieve this goal, a set of twenty-five lecture videos was taken from various departments at the University of Houston and Coursera website. The captions were produced with the assistance of the YouTube Speech Recognition System. The performances and limitations of OCR text, uncorrected/original speech text, and corrected speech text-based indexing was analyzed. The results indicate that slide text-based indexing yields 4% better results than spoken text-based indexing. The corrected speech text/caption provides better indexing results (11%) where OCR text fails to perform and the results closely matched the ground truth. The error analysis done on speech texts and slide texts prove that poor OCR text and caption quality are some of the main issues that hamper indexing accuracy.Item Peripheral Defocus and Myopia Progression in Myopic Children Randomly Assigned to Wear Single Vision and Progressive Addition Lenses(Investigative Ophthalmology & Visual Science, 2013-08) Berntsen, David A.; Barr, Christopher D.; Mutti, Donald O.; Zadnik, KarlaPurpose: To determine the effect of progressive addition lenses (PALs) and single vision lenses (SVLs) on peripheral defocus in myopic children, and to compare the effect of myopic versus hyperopic peripheral defocus on foveal myopia progression. Methods: Eighty-four myopic children aged 6 to 11 years with spherical equivalent (SE) cycloplegic autorefraction between −0.75 diopters (D) and −4.50 D were randomly assigned to wear SVLs or PALs. Aberrometry measurements of the eye and spectacles were made centrally, 30° nasally, temporally, and superiorly, and 20° inferiorly on the retina using a Complete Ophthalmic Analysis System for Vision Research (COAS-VR). The association between peripheral defocus and the 1-year change in central myopia was investigated. Results: SVLs caused a hyperopic shift in peripheral defocus at all locations (all P ≤ 0.0003). PALs caused a myopic shift in peripheral defocus in three of four locations measured (all P ≤ 0.01) with the greatest shift superiorly due to the PAL addition (−1.04 ± 0.30 D). Superior retinal defocus when wearing either SVLs or PALs was associated with the 1-year change in central myopia. The adjusted 1-year change in central SE myopia was −0.38 D for children with absolute superior myopic defocus (n = 67) and −0.65 D for children with absolute superior hyperopic defocus (n = 17; difference = 0.27 D; P = 0.002). Conclusions: PALs caused a myopic shift in peripheral defocus. Superior myopic defocus was associated with less central myopia progression. These data support the continued investigation of optical designs that result in peripheral myopic defocus as a potential way to slow myopia progression. (ClinicalTrials.gov number, NCT00335049.)Item Purpose in Life as a Tool for Managing Racial Discrimination and Suicide Ideation in Racial/Ethnic Minority Emerging Adults(2017-05) Hong, Judy H.; Walker, Rheeda L.; Vincent, John P.; Barr, Christopher D.An Existential-Constructivist approach to suicide resilience suggests that purpose in life (PIL) offsets one’s existential concerns, however, contextual challenges, such as racial discrimination, are believed to threaten both emotional wellbeing and contribute to risk for suicide. We posit that for racial and ethnic minority emerging adults (N=289) who perceive racial discrimination, having more PIL will be a more salient buffer than ethnic identity (EI) on thoughts of suicide. Regression analysis showed that EI was not a significant moderator for the relationship between perceived racial discrimination (PRD) on suicidal ideation (b = - .47, p = .13), however, PIL was a significant moderator (b = - .06, p = .025). A hierarchical regression showed that PIL as a moderator explained additional variance (∆R2 = 0.112, p < .001) in suicide ideation above and beyond EI as a moderator. Further investigation into the clinical significance of PIL as a buffer is necessary.