74 research outputs found
Deception detection in dialogues
No full textIn the social media era, it is commonplace to engage in written conversations. People sometimes even form connections across large distances, in writing. However, human communication is in large part non-verbal. This means it is now easier for people to hide their harmful intentions. At the same time, people can now get in touch with more people than ever before. This puts vulnerable groups at higher risk for malevolent interactions, such as bullying, trolling, or predatory behavior. Furthermore, such growing behaviors have most recently led to waves of fake news and a growing industry of deceit creators and deceit detectors. There is now an urgent need for both theory that explains deception and applications that automatically detect deception.
In this thesis I address this need with a novel application that learns from examples and detects deception reliably in natural-language dialogues. I formally define the problem of deception detection and identify several domains where it is useful. I introduce and evaluate new psycholinguistic features of deception in written dialogues for two datasets. My results shed light on the connection between language, deception, and perception. They also underline the challenges and difficulty of assessing perceptions from written text.
To automatically learn to detect deception I first introduce an expressive logical model and then present a probabilistic model that simplifies the first and is learnable from labeled examples. I introduce a belief-over-belief formalization, based on Kripke semantics and situation calculus. I use an observation model to describe how utterances are produced from the nested beliefs and intentions. This allows me to easily make inferences about these beliefs and intentions given utterances, without needing to explicitly represent perlocutions. The agents’ belief states are filtered with the observed utterances, resulting in an updated Kripke structure.
I then translate my formalization to a practical system that can learn from a small dataset and is able to perform well using very little structural background knowledge in the form of a relational dynamic Bayesian network structure.Submission published under a 24 month embargo labeled 'U of I Access', the embargo will last until 2019-08-01The student, Codruta Girlea, accepted the attached license on 2017-07-11 at 17:06.The student, Codruta Girlea, submitted this Dissertation for approval on 2017-07-11 at 17:12.This Dissertation was approved for publication on 2017-07-12 at 17:06.DSpace SAF Submission Ingestion Package generated from Vireo submission #11409 on 2018-03-02 at 13:01:34Made available in DSpace on 2018-03-02T19:59:34Z (GMT). No. of bitstreams: 3
GIRLEA-DISSERTATION-2017.pdf: 2591202 bytes, checksum: b93b631a28ce0708949fe3a67cf212d8 (MD5)
LICENSE.txt: 4211 bytes, checksum: 2d583705fbf074f8739b6ab5513a9c80 (MD5)
PROQUEST_LICENSE.txt: 4557 bytes, checksum: 577d1c5f0b0fdec857c9a83bafbcaf35 (MD5)
Previous issue date: 2017-07-12Embargo set by: Seth Robbins for item 105045
Lift date: 2020-03-02T19:59:52Z
Reason: Author requested U of Illinois access only (OA after 2yrs) in Vireo ETD systemEmbargo set by: Seth Robbins for item 105045
Lift date: 2020-03-02T20:02:46Z
Reason: Author requested U of Illinois access only (OA after 2yrs) in Vireo ETD systemU of I Only Restriction Lifted for Item 105045 on 2020-03-03T10:15:11Z
A Personalized Trial for Testing the Effects of a Mind-Body Intervention (MBI) on Sleep Duration and Quality in Middle-Aged Women Working in Health Care
No full textThis pilot study will utilize a Personalized Trials model to evaluate an individual participant’s experience with a Mind-Body Intervention (MBI) strategy for self-reported perceived stress and short sleep duration. Components of the MBI include mindfulness, yoga and guided walking, each assigned in 2-week block sequences for a total period of 12 weeks. Participants will be asked several questions a day sent via text message about their sleep quality, as well as bi-weekly questions about their stress, fatigue, concentration, confidence, mood, and pain levels to demonstrate relevant secondary impacts of sleep quality.
The study will start with a 2-week run-in period to assess baseline sleep duration and adherence to Fitbit wear and survey submission. After the 12 weeks of receiving the MBI, participants will enter a 2-week follow up period to allow for a personalized report of their observed data to be generated. Results from this pilot study will inform the future development of N-of-1 methodology in the research and clinical space aimed at addressing the health and wellness needs of adults biologically assigned female at birth who are 40-60 years of age
A Personalized Trial for Testing the Effects of a Mind-Body Intervention (MBI) on Sleep Duration and Quality in Middle-Aged Women Working in Health Care
No full textThis pilot study will utilize a Personalized Trials model to evaluate an individual participant’s experience with a Mind-Body Intervention (MBI) strategy for self-reported perceived stress and short sleep duration. Components of the MBI include mindfulness, yoga and guided walking, each assigned in 2-week block sequences for a total period of 12 weeks. Participants will be asked several questions a day sent via text message about their sleep quality, as well as bi-weekly questions about their stress, fatigue, concentration, confidence, mood, and pain levels to demonstrate relevant secondary impacts of sleep quality.
The study will start with a 2-week run-in period to assess baseline sleep duration and adherence to Fitbit wear and survey submission. After the 12 weeks of receiving the MBI, participants will enter a 2-week follow up period to allow for a personalized report of their observed data to be generated. Results from this pilot study will inform the future development of N-of-1 methodology in the research and clinical space aimed at addressing the health and wellness needs of adults biologically assigned female at birth who are 40-60 years of age
Recommended from our members
Design considerations for early-phase clinical trials of immune-oncology agents
No full textBackground
With numerous and fast approvals of different agents including immune checkpoint inhibitors, monoclonal antibodies, or chimeric antigen receptor (CAR) T-cell therapy, immunotherapy is now an established form of cancer treatment. These agents have demonstrated impressive clinical activity across many tumor types, but also revealed different toxicity profiles and mechanisms of action. The classic assumptions imposed by cytotoxic agents may no longer be applicable, requiring new strategies for dose selection and trial design.
Description
This main goal of this article is to summarize and highlight main challenges of early-phase study design of immunotherapies from a statistical perspective. We compared the underlying toxicity and efficacy assumptions of cytotoxic versus immune-oncology agents, proposed novel endpoints to be included in the dose-selection process, and reviewed design considerations to be considered for early-phase trials. When available, references to software and/or web-based applications were also provided to ease the implementation. Throughout the paper, concrete examples from completed (pembrolizumab, nivolumab) or ongoing trials were used to motivate the main ideas including recommendation of alternative designs.
Conclusion
Further advances in the effectiveness of cancer immunotherapies will require new approaches that include redefining the optimal dose to be carried forward in later phases, incorporating additional endpoints in the dose selection process (PK, PD, immune-based biomarkers), developing personalized biomarker profiles, or testing drug combination therapies to improve efficacy and reduce toxicity
The robotic approach significantly reduces length of stay after colectomy: a propensity score-matched analysis
No full textNaRnEA: An Information Theoretic Framework for Gene Set Analysis
No full textGene sets are being increasingly leveraged to make high-level biological inferences from transcriptomic data; however, existing gene set analysis methods rely on overly conservative, heuristic approaches for quantifying the statistical significance of gene set enrichment. We created Nonparametric analytical-Rank-based Enrichment Analysis (NaRnEA) to facilitate accurate and robust gene set analysis with an optimal null model derived using the information theoretic Principle of Maximum Entropy. By measuring the differential activity of ~2500 transcriptional regulatory proteins based on the differential expression of each protein’s transcriptional targets between primary tumors and normal tissue samples in three cohorts from The Cancer Genome Atlas (TCGA), we demonstrate that NaRnEA critically improves in two widely used gene set analysis methods: Gene Set Enrichment Analysis (GSEA) and analytical-Rank-based Enrichment Analysis (aREA). We show that the NaRnEA-inferred differential protein activity is significantly correlated with differential protein abundance inferred from independent, phenotype-matched mass spectrometry data in the Clinical Proteomic Tumor Analysis Consortium (CPTAC), confirming the statistical and biological accuracy of our approach. Additionally, our analysis crucially demonstrates that the sample-shuffling empirical null models leveraged by GSEA and aREA for gene set analysis are overly conservative, a shortcoming that is avoided by the newly developed Maximum Entropy analytical null model employed by NaRnEA
Considering Whether Medicaid Is Worth the Cost: Revisiting the Oregon Health Study
No full textThe Oregon Health Study was a groundbreaking experiment in which uninsured participants were randomized to either apply for Medicaid or stay with their current care.
The study showed that Medicaid produced numerous important socioeconomic and health benefits but had no statistically significant impact on hypertension, hypercholesterolemia, or diabetes. Medicaid opponents interpreted the findings to mean that Medicaid is not a worthwhile investment. Medicaid proponents viewed the experiment as statistically underpowered and, irrespective of the laboratory values, suggestive that Medicaid is a good investment.
We tested these competing claims and, using a sensitive joint test and statistical power analysis, confirmed that the Oregon Health Study did not improve laboratory values. However, we also found that Medicaid is a good value, with a cost of just $62 000 per quality-adjusted life-years gained
- …
