137 research outputs found
Bibliographics for the 983 eprints in the live archives of E-LIS : trends and status report up to 7th July 2004, based on author-self-archiving metadata
Get PDFThe priority for ideas and philosophy related to "Network Theory" have been traced back and documented by Braun(2004),and credit goes to Karinthy(1929).The IT has empowered to realise it, as the most practical phenomena and it is no more a humour. The OAI (Open Archives Initiatives)and ACIS (Academic Contributor Information System)are progressive in the direction ,which may lead to realise the "Collective Genius" at global level. Focus of present study is on Author-Self-Archiving (A-S-A)Metadata of the 983 Eprints in the Live Archives of the E-LIS (EPrints of Library and Information Science),which were approved till 7th July 2004.The A-S-A Metadata was used for librametric analysis. Self-explanatory bibliographics are illustrated.The highlights include: Conference papers (34%); highest approval, June 2004 (28%); published archives (76%);not refereed (52%); not in public domain (60%); highest self-archiving-author (De Robbio, Antonella).The Nos. of EPrints having single JITA domain specifications were: Theoretical and general aspects of libraries and information(27); Information use and sociology of information(80);Users,literacy and reading(13);Libraries as physical collections(30);Publishing and legal issues(57);Management(13);Industry, profession and education(36);Information sources, supports, channels(113) ; Information treatment for information services, Information functions and techniques (101); Technical services libraries, archives and museums(25); Housing technologies(1); Information technology and library technology(92); and Inter-domainery (395) i.e. having specifications of two or more than two JITA classes
Annual Report Assessment Using Large Language Models
No full textAnnual Report Assessment Using Large Language Models
When researching a given business, annual reports are among the most dependable and thorough resources available. When things seem to be going smoothly, annual reports should still be used to assist develop ideas and investment decisions, as well as identify red flags and early warning signals of problems.
The author of this study evaluated Bert and Roberta, two pre-trained language models for this job, and further fine-tuned them on Annual Reports. The author investigated zero-shot learning and few-shot learning using sentence-transformers for classification. The findings of the fine-tuned language models and further different classifiers are promising and will help financial analysts incorporate them into portfolio management and investment decision-making.
All work is based on Annual Report Assessment Dataset from the Author
DOI : 10.5281/zenodo.7536332The author is looking for feedback from the quantative investor community
Explainable AI for NLP: Decoding Black Box
No full textRecent advancements in machine learning have sparked greater interest in previously understudied topics. As
machine learning improves, experts are being pushed to understand and trace how algorithms get their results, how models
think, and why the end outcome. It is also difficult to communicate the outcome to end customers and internal stakeholders
such as sales and customer service without explaining the outcomes in simple language, especially using visualization. In
specialized domains like law and medicine, it becomes vital to understand the machine learning output.
The author talked about major categories, categorization, and techniques. Also, the author introduced a library called
NlpExplainer for the same. Currently, it works only for classification and shap. Future work involves incorporating more
tasks like QnA and summarization and implementing more techniques like LIME apart from Shap.The author introduced a library called NlpExplainer for XAI
Comparative study of efficacy and safety of pregabalin and gabapentin for treatment of neuropathic pain in oral cavity cancer patients: a comparative, randomized and prospective study
Get PDFBackground: Pain in patients with oral cancers can limit the normal functioning and quality of life. Neuropathic pain raises the anxiety and depression levels, increases the morbidity and decreases the efficiency to work. Neuropathic pain is frequently diagnosed as a complication of cancer pain due to direct invasion of nerves, plexus or compression, and side effect of chemotherapy, radiation injury or surgery.
Methods: A total of 60 patients were divided randomly into two groups based on treatment: group P (pregabalin) and group G (gabapentin). The intensity of pain was measured using visual analog scale (VAS) and DN4 questionnaire (Douleur Neuropathique 4) was used to evaluate neuropathic component. Changes in pain score and neuropathic component was assessed at 2nd and 4th week of follow up. Data was collected and analysed using SPSS 20.0 software at level of significance being p<0.05.
Results: At baseline, the mean±SD of VAS score in group P was 7.20±0.79; in group G was 7.13±0.66. At 2nd week, the mean±SD of VAS score in group P was 4.5±0.91; in group G was 4.46±0.88. At 4th week, the mean±SD of VAS score in group P was 3.66±0.69; in group G was 3.83±0.85. At baseline, the mean±SD of DN4 score in group P was 7.13±0.80; in group G was 6.93±0.85. At 2nd week, the mean±SD of DN4 score in group P was 4.73±0.92; in group G was 4.46±0.82. At 4th week, the mean±SD of DN4 score in group P was 3.73±0.42; in group G was 3.93±0.62.
Conclusions: Pregabalin was found to be more effective with lesser side effects than gabapentin
Synthesis of humidity sensitive zinc stannate nanomaterials and modelling of Freundlich adsorption isotherm model
No full text
Emvolio Instrument test reports and The ARRIVE guidelines 2.0 - Author checklist.pdf,
No full textThis also includes the details (1) test reports of Emvolio, as per the regulatory compliences (WHO/PQS/E003/TS01.1) and (2) Reports of Internal testing.
The file contains the ARRIVE2.0 Author Checklist for experiments carried out using rats.</p
Emerging 0D, 1D, 2D, and 3D nanostructures for efficient point-of-care biosensing
No full textThe recent COVID-19 infection outbreak has raised the demand for rapid, highly sensitive POC biosensing technology for intelligent health and wellness. In this direction, efforts are being made to explore high-performance nano-systems for developing novel sensing technologies capable of functioning at point-of-care (POC) applications for quick diagnosis, data acquisition, and disease management. A combination of nanostructures [i.e., 0D (nanoparticles & quantum dots), 1D (nanorods, nanofibers, nanopillars, & nanowires), 2D (nanosheets, nanoplates, nanopores) & 3D nanomaterials (nanocomposites and complex hierarchical structures)], biosensing prototype, and micro-electronics makes biosensing suitable for early diagnosis, detection & prevention of life-threatening diseases. However, a knowledge gap associated with the potential of 0D, 1D, 2D, and 3D nanostructures for the design and development of efficient POC sensing is yet to be explored carefully and critically. With this focus, this review highlights the latest engineered 0D, 1D, 2D, and 3D nanomaterials for developing next-generation miniaturized, portable POC biosensors development to achieve high sensitivity with potential integration with the internet of medical things (IoMT, for miniaturization and data collection, security, and sharing), artificial intelligence (AI, for desired analytics), etc. for better diagnosis and disease management at the personalized level
Combined experimental and molecular dynamics investigation of 1D rod-like asphaltene aggregation in toluene-hexane mixture
No full textThe aggregation behavior of asphaltene in the toluene-hexane mixture is systematically investigated using experimental techniques such as optical microscopy and Fourier-transform infrared spectroscopy (FTIR) combined with molecular dynamics (MD) simulations. The optical images of various asphaltene concentrations are processed to determine the size of asphaltene aggregates. FTIR is performed within our work in order to understand the formation of aggregates and its interaction with the solvent mixture. MD simulations are employed to achieve atomistic insights into the aggregation behavior of asphaltene. The end-to-end distance of the asphaltene molecule is calculated for various asphaltene concentrations in the toluene-hexane mixture. The dynamical properties of the asphaltene aggregates such as the diffusion coefficient and shear viscosity are calculated. Further, an in-depth analysis of the density contours is performed to probe the clusterization of asphaltene. Thus obtained structural and dynamical properties of the asphaltene aggregates in the toluene-hexane mixture are compared with our experimental findings. Our results thereby highlight the importance of the combined experimental and theoretical study to achieve deeper and better insights into the aggregation behavior of asphaltene in toluene-hexane mixture.Green Open Access added to TU Delft Institutional Repository 'You share, we take care!' - Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.Team Poulumi De
- …
