5153 research outputs found
Sort by
Deep Neural Language Model for Text Classification Based on Convolutional and Recurrent Neural Networks
No full textThe evolution of the social media and the e-commerce sites produces a massive amount of unstructured text data on the internet. Thus, there is a high demand to develop an intelligent model to process it and extract a useful information from it. Text classification plays an important task for many Natural Language Processing (NLP) applications such as, sentiment analysis, web search, spam filtering, and information retrieval, in which we need to assign single or multiple predefined categories to a sequence of text. In Neural Network Language Models learning long-term dependencies with gradient descent is difficult due to the vanishing gradient problem. Recently researchers started to increase the depth of the network in order to overcome the limitations of the existing techniques. However, increasing the depth of the network means increasing the number of the parameters, which makes the network computationally expensive, and more prone to overfitting. Furthermore, NLP systems traditionally treat words as discrete atomic symbols; the model can leverage small amounts of information regarding the relationship between the individual symbols. In recent years, deep learning models such as Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs) have been applied to language modeling with comparative, remarkable results. CNNs are a noble approach to extract higher-level features invariant to local translation. However, this method requires the stacking of multiple convolutional layers in order to capture long-term dependencies because of the locality of the convolutional and pooling layers. In this dissertation, we introduce a joint CNN-RNN framework to overcome the problems in the existing deep learning models. Briefly, we applied an unsupervised neural language model to train initial word embeddings that are further tuned by our deep learning network, then the pre-trained parameters of the network are used to initialize the model. At a final stage, the proposed framework combines former information with a set of feature maps learned by a convolutional layer with long-term dependencies learned via Long-Short-Term Memory (LSTM). Empirically, we show that our approach, with slight hyperparameter tuning and static vectors, achieves outstanding results on multiple sentiment analysis benchmarks. Our approach outperforms several existing approaches in term of accuracy; our results are also competitive with the state-of-the-art results on the Stanford Large Movie Review (IMDB) dataset, and the Stanford Sentiment Treebank (SSTb) dataset. Our approach has a significant role in reducing the number of parameters and constructing the convolutional layer followed by the recurrent layer with no pooling layers. Our results show that we were able to reduce the loss of detailed, local information and capture long-term dependencies with an efficient framework that has fewer parameters and a high level of performance
The Power of Desire in American Self-imagining Or, Thomas Jefferson’s Lost Trunk
No full textDiane Krumrey's poster on Thomas Jefferson's empty trunk and positivist view on universal language and American language and it's influence on later thinkers
Demographic and Outcome Trends in a Naturopathic Teaching Clinic
No full textNaturopathic Doctors (NDs) are trained to treat patients using an individualized approach combining dietary and lifestyle changes, botanical and nutritional medicines, homeopathic medicines, and counseling techniques. In order to effectively study naturopathic medicine, it is beneficial to use whole systems research rather than standardized protocols. In our study, we monitored 27 patients treated individually with wholistic naturopathic medicine for various chief concerns over the course of 12 weeks in the UB Naturopathic Clinic. These patients were given surveys at baseline, 6 weeks, and 12 weeks pertaining to their health-related quality of life (HRQoL), which is considered by the CDC to be an effective tool for predicting mortality and morbidity compared to many objective measures. We used the PROMIS Global Health Scale questionnaire, a validated tool for monitoring HRQoL, for these measures. We also collected demographic data at baseline, including gender, chief medical concerns, age, ethnicity, income level, and education. In our study population, 38.5% of patients were seeking care for depression/anxiety and 15.4% of patients were seeking care for fatigue. 46.2% of participants are aged 50-69 years old, and 71.8% are female. 28.2% have an income under $29,999 per year, and 51.3% are Caucasian. PROMIS Global Health Measures are divided into Physical Health and Mental Health T-scores, and a T-score of 50 is considered an average score for a control population sampled by PROMIS. GPH scores started at 44.652 and increased to 47.8 at 6 weeks and declined to 45.324 at 12 weeks (average SD +/- 7.94). GMH scores started at 45.5259 at baseline, increased to 46.5185 at 6 weeks, and increased to 47.356 at 12 weeks (average SD +/- 8.40). While not statistically significant, our research helps us to understand the type of patients who visit a naturopathic clinic and to perform future research to target treatments for this particular population. A trend towards improved GMH scores was noted over 12 weeks
MQTT and ROC Based Hybrid Robot as a Service (RaaS) Platform
No full textRobots are rapidly evolving from factory work-horses to robot-companions. The future of robots will be as companions in the workplace functioning as interactive salespeople. In order to support this transition, it is important to combine service-oriented architecture and robotics. Service-oriented architecture and cloud computing have become dominant computing paradigms, and adding an RaaS (Robot as a Service) unit as a part of this system will help the companies manage and develop robots more efficiently. The major components of RaaS will be the integration of RMS (Robot Management System) and ROC (Robot Operation Center). As more and more robots are increasing in the service industry, the inter-robot communication is very critical. This communication can be achieved by ROC and the robots can be monitored remotely or locally via RMS. The RaaS platform will comply with all the standards of SOA (Service Oriented Architecture) like the development platform and execution unit, thereby creating a flexible and more development-friendly process
Mere Exposure Effect and Its Application in Business
No full textThe mere-exposure effect is a psychological phenomenon that people more tend to choose or hold a preference for the things, which they are familiar with. This report devotes to investigate this physiology bias by two designed survey questions
Understanding and Treating Posttraumatic Stress Disorder Using the Six Principles of Naturopathic Medicine
No full textPosttraumatic Stress Disorder (PTSD) is a health condition that has been increasing in occurrence due to a high number of traumatic events individuals have been experiencing around the world as well as the awareness of the effects of such traumatic events on individuals. There are several effective treatment options available for those who have PTSD. This includes conventional therapies, as well as complementary and alternative medicine approaches. Naturopathic Medicine is one field of medicine that can be utilized by patients with PTSD seeking safe and effective treatments. A complete understanding of PTSD including risk factors, symptoms and neurophysiology is important in determining an appropriate treatment plan for patients. Knowing that each case presentation may be different will play a role in providing the most effective individualized treatment plan. With the increased understanding of PTSD, there are many ways to approach a PTSD case, one of which can involve the use of the six principles of Naturopathic Medicine. This approach uses a variety of modalities as part of an effective treatment plan. It will take into account the individuality of each case as well as constitute a holistic approach to each presenting PTSD patient. This thesis/poster aims to link the six principles of Naturopathic Medicine to aspects of PTSD that allows for us to understand and potentially treat PTSD
