22505 research outputs found
Sort by
Pedagogy and Practice in the Online Creative Writing Classroom
Get PDFReview of Tamara Girardi and Abigail G. Scheg, editors. Theories and Strategies for Teaching Creative Writing Online. Routledge, 2021. 165 pages
Fake News Detection: Leveraging Natural Language Processing and Machine Learning for Reliable Information Verification
Get PDFThis dissertation details the creation and assessment of a machine learning algorithm designed to identify fake news utilizing Natural Language Processing (NLP) methods. The research employs several machines learning models, including Long Short-Term Memory (LSTM) and other deep learning techniques, to detect and classify misleading information. Data is sourced from a variety of platforms, such as social media and online news outlets, to compile a thorough dataset. The data is pre-processed to eliminate noise, address missing values, and extract essential features through techniques like tokenization, stop-word removal, and lemmatization. The performance of the models is evaluated using key metrics such as accuracy, precision, recall, and F1-score. The results indicate that LSTM models surpass traditional methods, offering more precise and trustworthy fake news detection. Additionally, the research investigates hybrid models that integrate multiple machine learning strategies to enhance classification accuracy. These results underscore the promise of AI-driven fake news detection systems in addressing misinformation, especially in political settings, while also demonstrating their usefulness for real-time content filtering on social media platforms
Flow dynamics of agricultural waste nanofibers: shear, temperature, and oscillatory insights
Get PDFThe rheology and fiber size of corn stover (CS) and cleaned cotton gin trash (CGT) cellulose nanofibers (CNFs) were studied including behaviors at both moderate and extremely high shear rates, to simulate industrial processes ranging from mixing and pumping to high-speed coating, printing, and extrusion. Particle size analyzer showed that 99% of CS fibers measured around 226 nm, while 85% of CGT fibers were approximately 143 nm. Both CS and CGT CNFs formed gel-like suspensions, and shear flow tests revealed that all samples exhibited shear-thinning behavior, allowing easy flow under shear forces. Gels with higher solid content (1%) demonstrated extended viscoelastic regions, indicating improved flexibility, structural integrity, and energy-dispersing capacity. The combination of shear-thinning behavior, flexibility, and energy absorption makes these CNFs promising for applications in papermaking, coatings, packaging, and areas where shock absorption is required
Defining transcriptional signatures of susceptibility for ceftriaxone and tetracycline in Neisseria gonorrhoeae
Get PDFResistance to antibiotics is the ability of bacteria to survive in spite of the administration of drugs designated to kill them; and overuse of antibiotics over the past decades has led to a vast increase in resistance. Most infections are treated empirically, such that patients are given a recommended course of antibiotics, however these infections are not frequently tested for their antimicrobial susceptibility profiles (i.e., tested only in the case of treatment failure). This can lead to inappropriate prescription which can increase the selective pressure of antibiotics on other bacterial members of our microbiomes. Susceptibility testing would help determine the antibiotics that are the most effective in helping the prompt treatment of disease-causing organism. In this study we nominated biological signatures to serve as the foundation for developing a point-of-care (POC) diagnostic tool for antimicrobial susceptibility testing. RNA transcriptional signatures of Neisseria gonorrhoeae were generated upon drug exposure, as RNA abundance change can define susceptibility; susceptible bacteria are expected to express a stress response after drug exposure, which is absent in resistant bacteria. These profiles were then analyzed to identify Treatment-Susceptibility (TS) patterns; those transcripts differentially expressed in susceptible cells across treatment, and differentially expressed in the treatment condition. The transcripts exhibiting distinct TS pattern were investigated for each drug contrast and at two specific time points i.e., 120 minutes and 240 minutes. We identify 61 transcripts that show the TS pattern for ceftriaxone at 120 minutes and 7 at 240; and 18 transcripts that show the TS pattern for tetracycline at 120 minutes and 128 at 240. The results indicate that RNA abundance changes can reliably signify bacterial susceptibility, as susceptible isolates demonstrate a clear transcriptional response to drug exposure. This study also demonstrated that to acquire meaningful results it is crucial to test transcripts at specific timepoints, as they capture the dynamic transcriptional response of bacteria to drug exposure. By nominating these biological signatures, this study lays the groundwork for developing a rapid, RNA- based diagnostic tool. Such a tool could enhance susceptibility testing, reduce inappropriate antibiotic use, and support more targeted treatment strategies. Overall, these findings depicted the most useful transcriptional signatures, which could serve as a key component in future point-of-care (POC) devices for antimicrobial susceptibility testing
SoundHapticsVR: Head-Based Spatial Haptic Feedback for Accessible Sounds in Virtual Reality for Deaf and Hard of Hearing Users
Get PDFVirtual Reality (VR) systems use immersive spatial audio that provide more localized sound awareness to convey critical information. However, these audio cues are often inaccessible to individuals who are Deaf or Hard-of-Hearing (DHH). To address this, we developed, a head-based haptic system designed to enable localizing and convert- ing audio signals directly into haptic feedback using multi-channel acoustic haptic actuators. We evaluated through three studies. The preliminary study determined the maximum tactile frequency threshold perceivable on different regions of the head of DHH users when a VR headset is attached. This was followed by a study to deter- mine the ideal number and arrangement of transducers for effective sound localization on the head for VR applications. Our third study focused on differentiating sound sources, where participants associated spatial sound sources with corresponding spa- tial and acoustic haptic patterns. Our findings indicate that tactile perception thresh- olds vary across different head regions, emphasizing consistent frequency equalization across all position to ensure uniform sensation. Adding a transducer to the front of the head significantly improved sound localization, serving as a critical confirmation cue for DHH users. Participants were able to correlate distinct haptic patterns with specific objects, suggesting that unique and more intense haptic patterns were helpful in localizing the contextual sound source. Overall, the system has the potential to make VR applications that has wide range of sound information to be more accessible for DHH users
Impact of Precursor Solvent Chemistry on Electrical Properties of Indium Oxide Films for Thin-Film Transistors
Get PDFIndium oxide In_2O_3 combines high carrier mobility with optical transparency, making it a promising candidate for next-generation thin-film electronics. However, its electrical performance is highly sensitive to fabrication parameters, particularly the chemistry of solution-based precursors, which can alter defect landscapes and carrier transport mechanisms. This work systematically examines how solvent choice influences the structural and electronic properties of solution-processed In_2O_3 thin films, isolating processing–property relationships that remain insufficiently understood. Films prepared from water and 2-methoxyethanol (2ME) precursors were characterized using temperature-dependent transmission line method (TLM) measurements from 30~K to 300~K. Water-processed films exhibited consistently lower sheet resistance, a reduced activation energy (2.0~meV vs.\ 12~meV for 2ME), and a significantly lower hopping parameter T_0, indicating reduced spatial and energetic disorder. AFM revealed smoother surface morphology for the water-based films, while XPS showed a higher oxygen vacancy-to-lattice oxygen ratio, consistent with enhanced carrier density. These results demonstrate that solvent identity can be used to tune both functional and structural disorder, enabling substantial improvements in charge transport without doping or high-temperature processing. This solvent-driven approach provides a viable pathway for engineering high-performance, low-temperature oxide semiconductors for applications such as radiation-tolerant and flexible electronics
Protein Language Models-Based Representation for Post-translational Modification Prediction
Get PDFPost-translational modifications (PTMs) are chemical changes that occur after translation and play a key role in regulating protein function and cellular processes. Their dysregulation is associated with various diseases, making accurate prediction of PTM sites essential for understanding cellular mechanisms and informing therapeutic strategies. This work focuses on the computational prediction of two biologically significant PTMs, namely succinylation and O-GlcNAcylation, utilizing recent advances in protein language models (pLMs). Drawing an analogy to natural languages, amino acids are treated as words and sequences as sentences, allowing pLMs to capture contextual dependencies within protein sequences. One of the earliest contributions of this thesis, LMSuccSite, is a framework that integrates both global and local contextual information from protein sequences to improve the identification of succinylation sites in proteins. The global context is captured using pretrained pLMs, which learn semantic representations of amino acid sequences by modeling dependencies across distant residues. To complement this, the local context is incorporated through a supervised word embedding model that encodes dense representations of fixed-length sequence windows centered around target residues. Global and local contextual representations are subsequently fused to produce a unified representation that benefits from both broad sequence-level understanding and fine-grained residue-level patterns. This fusion addresses a key limitation of prior approaches, which are based on handcrafted features that are often restricted to local context. By integrating embeddings from both perspectives, LMSuccSite offers a more comprehensive approach to succinylation prediction. Building on the effectiveness of protein language models in succinylation site prediction, the second project in this thesis focuses on identifying O-GlcNAcylation sites using a model named LM-OGlcNAc-Site, where we explored various strategies to integrate the multiple sequence-based protein language models. LM-OGlcNAc-Site integrates multiple pLMs with diverse architectures and training objectives, as each captures distinct contextual representations of protein sequences. Their fusion enables a more comprehensive and informative representation of proteins. This integrated approach improved the identification of O-GlcNAc sites compared to individual representation, demonstrating the effectiveness in enhancing PTM site prediction. Expanding upon the sequence-based protein representations, we further investigated the integration of structure-aware pLMs for residue-level and protein-level classification tasks, aiming to capture complementary information embedded in protein sequences and their 3D conformations. Several state-of-the-art sequence-based and structure-aware pLMs were systematically evaluated to extract high-dimensional embeddings, and multiple fusion strategies were explored to integrate these diverse feature sets effectively. While sequence-based models demonstrated strong baseline performance, incorporating structural context consistently yielded competitive results in most of the experiments. These experiments highlight the potential of incorporating structural information to enhance protein function prediction