Search

Your search keyword '"Nawshad A"' showing total 2,387 results
Start Over Author "Nawshad A"
2,387 results on '"Nawshad A"'

1. Quad core gold coated photonic crystal fiber temperature sensor based on surface plasmon resonance

Author

MD. Shahriar Karim, Sazzad Hossin, Md. Rafiul Alam, Md. Abu Bakar Siddik, Mst. Rubina Aktar, Nawshad Ahmed, and Md. Abdullah Noman Shakh

Subjects
Quad core photonic crystal fiber, Finite element method (FEM), Temperature sensor, Surface plasmon resonance (SPR), Plasmonic material, Temperature sensitive liquid, Engineering (General). Civil engineering (General), TA1-2040
Abstract

This paper shows a simple quad core Photonic Crystal Fiber (PCF) temperature sensor based on Surface Plasmon Resonance (SPR) which has very high sensitivity and sensor resolution. For evaluating the sensing performance Finite Element Method (FEM) is used. In this design, gold is used as plasmonic material and the outer layer of the designed PCF has to be filled by it for making the fabrication easier than any other model. Besides, the Chloroform is used as the temperature sensitive liquid (analyte). The designed temperature sensor has the maximum wavelength sensitivities which are 7.5 nm/°C and 10.5 nm/°C for x and y-polarization respectively. The sensor shows the resolution of 13.33 × 10−3 RIU for x-polarization and 9.52 × 10−3 RIU for y-polarization. Core diameter, radius of air hole, thickness of gold and pitch are the structural parameters which are adjusted to their optimized value for getting higher sensitivity. Easy fabrication and simplicity are the major aspects of this proposed model which makes it applicable to various applications such as medical science, power plant, engineering sector, environment monitoring etc.

Published
2023
Full Text
View/download PDF

2. BUET Multi-disease Heart Sound Dataset: A Comprehensive Auscultation Dataset for Developing Computer-Aided Diagnostic Systems

Author

Ali, Shams Nafisa, Zahin, Afia, Shuvo, Samiul Based, Nizam, Nusrat Binta, Nuhash, Shoyad Ibn Sabur Khan, Razin, Sayeed Sajjad, Sani, S. M. Sakeef, Rahman, Farihin, Nizam, Nawshad Binta, Azam, Farhat Binte, Hossen, Rakib, Ohab, Sumaiya, Noor, Nawsabah, and Hasan, Taufiq

Subjects
Electrical Engineering and Systems Science - Signal Processing, Computer Science - Artificial Intelligence, Computer Science - Machine Learning, Computer Science - Sound, Electrical Engineering and Systems Science - Audio and Speech Processing
Abstract

Cardiac auscultation, an integral tool in diagnosing cardiovascular diseases (CVDs), often relies on the subjective interpretation of clinicians, presenting a limitation in consistency and accuracy. Addressing this, we introduce the BUET Multi-disease Heart Sound (BMD-HS) dataset - a comprehensive and meticulously curated collection of heart sound recordings. This dataset, encompassing 864 recordings across five distinct classes of common heart sounds, represents a broad spectrum of valvular heart diseases, with a focus on diagnostically challenging cases. The standout feature of the BMD-HS dataset is its innovative multi-label annotation system, which captures a diverse range of diseases and unique disease states. This system significantly enhances the dataset's utility for developing advanced machine learning models in automated heart sound classification and diagnosis. By bridging the gap between traditional auscultation practices and contemporary data-driven diagnostic methods, the BMD-HS dataset is poised to revolutionize CVD diagnosis and management, providing an invaluable resource for the advancement of cardiac health research. The dataset is publicly available at this link: https://github.com/mHealthBuet/BMD-HS-Dataset., Comment: 14 pages, 13 figures

Published
2024

8. Deep Temporal Modelling of Clinical Depression through Social Media Text

Author

Farruque, Nawshad, Goebel, Randy, Sivapalan, Sudhakar, and Zaïane, Osmar R.

Subjects
Computer Science - Computation and Language, Computer Science - Machine Learning
Abstract

We describe the development of a model to detect user-level clinical depression based on a user's temporal social media posts. Our model uses a Depression Symptoms Detection (DSD) classifier, which is trained on the largest existing samples of clinician annotated tweets for clinical depression symptoms. We subsequently use our DSD model to extract clinically relevant features, e.g., depression scores and their consequent temporal patterns, as well as user posting activity patterns, e.g., quantifying their ``no activity'' or ``silence.'' Furthermore, to evaluate the efficacy of these extracted features, we create three kinds of datasets including a test dataset, from two existing well-known benchmark datasets for user-level depression detection. We then provide accuracy measures based on single features, baseline features and feature ablation tests, at several different levels of temporal granularity. The relevant data distributions and clinical depression detection related settings can be exploited to draw a complete picture of the impact of different features across our created datasets. Finally, we show that, in general, only semantic oriented representation models perform well. However, clinical features may enhance overall performance provided that the training and testing distribution is similar, and there is more data in a user's timeline. The consequence is that the predictive capability of depression scores increase significantly while used in a more sensitive clinical depression detection settings., Comment: Tables are properly oriented and some more typos were fixed

Published
2022

9. Depression Symptoms Modelling from Social Media Text: A Semi-supervised Learning Approach

Author

Farruque, Nawshad, Goebel, Randy, Sivapalan, Sudhakar, and Zaiane, Osmar

Subjects
Computer Science - Computation and Language, Computer Science - Artificial Intelligence, Computer Science - Machine Learning
Abstract

A fundamental component of user-level social media language based clinical depression modelling is depression symptoms detection (DSD). Unfortunately, there does not exist any DSD dataset that reflects both the clinical insights and the distribution of depression symptoms from the samples of self-disclosed depressed population. In our work, we describe a Semi-supervised Learning (SSL) framework which uses an initial supervised learning model that leverages 1) a state-of-the-art large mental health forum text pre-trained language model further fine-tuned on a clinician annotated DSD dataset, 2) a Zero-Shot learning model for DSD, and couples them together to harvest depression symptoms related samples from our large self-curated Depression Tweets Repository (DTR). Our clinician annotated dataset is the largest of its kind. Furthermore, DTR is created from the samples of tweets in self-disclosed depressed users Twitter timeline from two datasets, including one of the largest benchmark datasets for user-level depression detection from Twitter. This further helps preserve the depression symptoms distribution of self-disclosed Twitter users tweets. Subsequently, we iteratively retrain our initial DSD model with the harvested data. We discuss the stopping criteria and limitations of this SSL process, and elaborate the underlying constructs which play a vital role in the overall SSL process. We show that we can produce a final dataset which is the largest of its kind. Furthermore, a DSD and a Depression Post Detection (DPD) model trained on it achieves significantly better accuracy than their initial version., Comment: Title and relevant changes are made

Published
2022

14. Mesoporous silica-grafted deep eutectic solvent-based mixed matrix membranes for wastewater treatment: Synthesis and emerging pollutant removal performance

Author

Rashid Kanwal, Aslam Muhammad, Rácz Ervin, Nadeem Sohail, Khan Zakir, Muhammad Nawshad, Rashid Zeeshan, Aljuwayid Ahmed M., Shahid Muhammad Kashif, and Irfan Muhammad

Subjects
mesoporous silica, membrane, deep eutectic solvent, heavy metal, wastewater, Technology, Chemical technology, TP1-1185, Physical and theoretical chemistry, QD450-801
Published
2024
Full Text
View/download PDF

15. Assessment of chlorohexidine-coated cotton filler addition on resin-based fissure sealants’ mechanical, water sorption, and antibacterial properties

Author

Sajida Gul, Muhammad Adnan Khan, Saad Liaqat, Naveed Ahmad, and Nawshad Muhammad

Subjects
Cotton cellulose, chlorhexidine, pits and fissures sealants, mechanical properties, water sorption, antibacterial properties, Materials of engineering and construction. Mechanics of materials, TA401-492
Abstract

AbstractThe occlusal surfaces of posterior teeth are susceptible to caries formation due to the presence of pits and fissures. The bacteria present in the oral cavity and food particles are accumulated on these complex structures. To prevent and protect the tooth structure, different additives were added to pits and fissures sealants to improve adhesion with tooth structure, enhance the mechanical properties, and prevent bacterial growth. In this regard, cotton was coated with chlorohexidine (CHX) and incorporated in a resin-based sealant to improve its mechanical properties (compressive and flexural strength), mass change, and bacterial properties. The mean compressive strength values of CHX coated cotton modified RBS samples S1, S2 & S3 were significantly higher than the control group. The mean flexural strength value of group S1 was significantly higher as compared to the C, S2 & S3. Disc diffusion methods showed a significantly higher value of (p-value = 0.003) antibacterial potential of groups S2 and S3 as compared to group C. The mean percent mass change values of S1, S2 & S3 were higher than the control group. Hence the addition of CHX-coated cotton fillers in RBS improved its properties.

Published
2024
Full Text
View/download PDF

16. Contributors

Author

Abdin, Zainul, primary, Afzal, Mahvash, additional, Aghakhani, Arash, additional, Aisbett, Emma, additional, Burolia, Akash Kumar, additional, Catchpole, Kylie, additional, Chandrappa, Sujana, additional, Erkovan, Mustafa, additional, Galbao, Simon Joyson, additional, Ghosh, Biswajit, additional, Gourlay, David, additional, Guzzini, Alessandro, additional, Haque, Nawshad, additional, Jaju, Santosh, additional, Kilinc, Necmettin, additional, Kshirsagar, Soham, additional, Kumar, Abhishek, additional, Kumar, Sanjay, additional, Liu, Yun, additional, Lv, Peng, additional, Mahajan, Dhiraj K., additional, Mandal, Ananya, additional, Manna, Joydev, additional, Mukhopadhyay, Nilay Krishna, additional, Murthy, Dharmapura H.  K., additional, Nallapaneni, Anuraag, additional, Nanan-Surujbally, Anilia, additional, Neogi, Swati, additional, Pathak, Dinesh, additional, Pellegrini, Marco, additional, Saccani, Cesare, additional, Shaneeth, M, additional, Sharma, Pranjali, additional, Sharma, Davinder Pal, additional, Sisman, Orhan, additional, Sontakke, Ujwal R., additional, Subramaniam, Anandh, additional, Tang, Chunguang, additional, White, Lee V., additional, and Yadav, Thakur Prasad, additional

Published
2024
Full Text
View/download PDF

17. Clinicopathological discrepancies in the diagnoses of childhood causes of death in the CHAMPS network: An analysis of antemortem diagnostic inaccuracies

Author

Khatia Munguambe, Quique Bassat, Robert F Breiman, Mahbubur Rahman, Sibone Mocumbi, Cynthia G Whitney, Amy Wise, Shams El Arifeen, Meerjady Sabrina Flora, Soter Ameh, Emily S Gurley, Mustafizur Rahman, Ariel Nhacolo, Carla Carrilho, Cheick Bougadari Traore, Lucy Liu, Inacio Mandomando, Clara Menendez, Chris A Rees, Janet Agaya, Jane Juma, Tadesse Gure, George Aol, Hennie Lombaard, Ziyaad Dangor, Shabir Madhi, James Bunn, Samba O Sow, Dickson Gethi, Sanwarul Bari, Jeffrey P Koplan, Natalia Rakislova, Fabiola Fernandes, Tacilta Nhampossa, Maria Maixenchs, Mohammed Kamal, Nega Assefa, Joseph Oundo, Lola Madrid, Tahmina Shirin, Nelesh P Govender, Addisu Alemu, Hailemariam Legesse, Uma U Onwuchekwa, Awa Traore, Portia C Mutevedzi, Victor Akelo, Sammy Khagayi, Dickens Onyango, Richard Omore, Yasmin Adam, Peter Otieno, Margaret Mannah, Vicky Baillie, Karen L Kotloff, John Blevins, Milagritos D Tapia, Rima Koka, Mohammad Zahid Hossain, Dickens Kowuor, Tom Sesay, James Squire, Francis Moses, Haleluya Leulseged, Christine Bethencourt, Kitiezo Aggrey Igunza, Ikechukwu U Ogbuanu, Andrew Moseray, Ima-Abasi Bassey, Afruna Rahman, Muntasir Alam, Yasir Y Abdullahi, Nana Kourouma, Seydou Sissoko, Rosauro Varo, Sana Mahtab, Martin Hale, Jeanie du Toit, Zachary J Madewell, Dianna M Blau, Roosecelis B Martines, Fatima Solomon, Gillian Sorour, Jeannette Wadula, Karen Petersen, Peter J Swart, Sanjay G Lala, Sithembiso Velaphi, Richard Chawana, Ashleigh Fritz, Nellie Myburgh, Pedzisai Ndagurwa, Cleopas Hwinya, Shahana Parveen, ASM Nawshad Uddin Ahmed, Mahbubul Hoque, Saria Tasnim, Ferdousi Islam, Farida Ariuman, Mohammad Mosiur Rahman, Ferdousi Begum K Zaman, Dilruba Ahmed, Alexander M Ibrahim, Fikremelekot Temesgen, Melisachew Mulatu Yeshi, Mahlet Abayneh Gizaw, Stian MS Orlien, Solomon Ali, Peter Nyamthimba Onyango, Richard Oliech, Joyce Akinyi Were, Thomas Misore, Harun Owuor, Christopher Muga, Bernard Oluoch, Christine Ochola, Sharon M Tennant, Carol L Greene, Ashka Mehta, J Kristie Johnson, Brigitte Gaume, Adama Mamby Keita, Karen D Fairchild, Diakaridia Kone, Diakaridia Sidibe, Doh Sanogo, Kounandji Diarra, Tiéman Diarra, Kiranpreet Chawla, Zara Manhique, Fatmata Bintu Tarawally, Martin Seppeh, Ronald Mash, Julius Ojulong, Babatunde Duduyemi, Alim Swaray-Deen, Joseph Bangura Amara Jambai, Okokon Ita, Cornell Chukwuegbo, Sulaiman Sannoh, Princewill Nwajiobi, Erick Kaluma, Oluseyi Balogun, Carrie Jo Cain, Solomon Samura, Samuel Pratt, Joseph Kamanda Sesay, Osman Kaykay, Binyam Halu, Francis Smart, Sartie Kenneh, Jana Ritter, Tais Wilson, Jonas Winchell, Jakob Witherbee, Navit T Salzberg, Margaret Basket, Ashutosh Wadhwa, Kyu Han Lee, Roosecelis Martines, Shamta Warang, Maureen Diaz, Jessica Waller, Shailesh Nair, Courtney Bursuc, Kristin LaHatte, Sarah Raymer, Solveig Argeseanu, Kurt Vyas, and Manu Bhandari

Subjects
Pediatrics, RJ1-570
Abstract

Introduction Determining aetiology of severe illness can be difficult, especially in settings with limited diagnostic resources, yet critical for providing life-saving care. Our objective was to describe the accuracy of antemortem clinical diagnoses in young children in high-mortality settings, compared with results of specific postmortem diagnoses obtained from Child Health and Mortality Prevention Surveillance (CHAMPS).Methods We analysed data collected during 2016–2022 from seven sites in Africa and South Asia. We compared antemortem clinical diagnoses from clinical records to a reference standard of postmortem diagnoses determined by expert panels at each site who reviewed the results of histopathological and microbiological testing of tissue, blood, and cerebrospinal fluid. We calculated test characteristics and 95% CIs of antemortem clinical diagnostic accuracy for the 10 most common causes of death. We classified diagnostic discrepancies as major and minor, per Goldman criteria later modified by Battle.Results CHAMPS enrolled 1454 deceased young children aged 1–59 months during the study period; 881 had available clinical records and were analysed. The median age at death was 11 months (IQR 4–21 months) and 47.3% (n=417) were female. We identified a clinicopathological discrepancy in 39.5% (n=348) of deaths; 82.3% of diagnostic errors were major. The sensitivity of clinician antemortem diagnosis ranged from 26% (95% CI 14.6% to 40.3%) for non-infectious respiratory diseases (eg, aspiration pneumonia, interstitial lung disease, etc) to 82.2% (95% CI 72.7% to 89.5%) for diarrhoeal diseases. Antemortem clinical diagnostic specificity ranged from 75.2% (95% CI 72.1% to 78.2%) for diarrhoeal diseases to 99.0% (95% CI 98.1% to 99.6%) for HIV.Conclusions Antemortem clinical diagnostic errors were common for young children who died in areas with high childhood mortality rates. To further reduce childhood mortality in resource-limited settings, there is an urgent need to improve antemortem diagnostic capability through advances in the availability of diagnostic testing and clinical skills.

Published
2024
Full Text
View/download PDF

18. Evaluation of physicochemical properties of Rohu fish noodles

Author

Sabrina Zaman, Fazlur Rahman, Mahfujul Alam, Md. Numan Islam, Rashida Parvin, Nazia Nawshad Lina, and Md. Ashrafuzzaman Zahid

Subjects
Noodles, Fish noodles, Value-added fish product, Rohu fish, Physicochemical properties, Food processing and manufacture, TP368-456
Abstract

Functional noodles are considered nutritious and healthy staple food products. This study aimed to develop value-added rohu fish noodles (Labeo rohita) and evaluate their physicochemical properties. Six noodle samples were prepared following several formulations. After a 2-day drying process, the noodles were stored in air-resistant plastic bags. The formulated noodles were investigated based on their physical (water absorption capacity, cooking loss, and volume increase) and chemical (moisture, ash, protein, fat, and fiber) properties. Statistical analysis revealed that the inclusion of rohu fish mince significantly increased moisture and protein content (p0.05) were found in ash, fiber, fat, and mineral content among the various fish noodle samples. This study concludes that the incorporation of rohu fish mince can enhance the nutritional properties of traditional noodles.

Published
2024
Full Text
View/download PDF

19. Techno-economic evaluation of an environmental-friendly processing route to extract rare earth elements from monazite

Author

Reiza Z. Mukhlis, Jin-Young Lee, Hee Nam Kang, Nawshad Haque, Mark I. Pownceby, Warren J. Bruckard, M. Akbar Rhamdhani, and Rajesh Kumar Jyothi

Subjects
Monazite, Rare earth elements, Critical mineral, Critical metals, Technoeconomic analysis, Renewable energy sources, TJ807-830, Environmental engineering, TA170-171
Abstract

The relatively high concentration of thorium and uranium in monazite poses significant environmental issues in the extraction process of rare earth elements from monazite if left untreated. A new process route that recovers both thorium and uranium as oxides has been proposed, and an analysis of its economic feasibility is presented in the current paper, based on analysis and processing of a Korean monazite sample. Comparative evaluations with existing acidic and alkaline routes have also been carried out. It was estimated that the largest proportion of the operational cost for the new process related to the materials and reagent costs. Sensitivity analysis predicted that the value of neodymium oxide followed by Heavy Rare Earth Oxides (HREO) and praseodymium oxide affect the revenue significantly. Increasing the average basket price of the total rare earths oxide by 1.5 times would result in revenues of US$158.8 million/year for the proposed route, compared to US$161.5 and US$156.3 million/year for the alkaline and acidic routes, respectively. The discounted cash flow analysis and the resulted Net Present Value (NPV) suggested that the proposed processing route was in fact the only process estimated to be economically feasible with the payback period expected to be around 4.5 years. The sale of thorium oxide and uranium oxide by-products of the proposed route contributed to the positive discounted NPV. It was also estimated that a minimum sale price of US$20/kg total rare earth oxide is required to ensure all the processes generate a positive discounted NPV. These results shown that the proposed new processing route is estimated to be economically feasible.

Published
2024
Full Text
View/download PDF

22. Clinical characteristics and oral manifestations of COVID-19: A cross-sectional study from Peshawar, Pakistan

Author

Sana Hussain, Maryam Rauf, Saima Aleem, Khalid Rehman, Muhammad Amer Khan, Saad Liaqat, Nawshad Muhammad, and Muhammad adnan Khan

Subjects
dysgeusia, covid-19, oral manifestations, taste disorders, xerostomia, oral ulcer, anosmia, fever, fatigue, cough, Medicine
Abstract

OBJECTIVES: To determine the prevalence of oral manifestations among PCR-confirmed COVID-19-positive patients and their potential correlations with age, gender, and systemic manifestations. METHODS: This cross-sectional study was conducted at Khyber Medical University among 117 randomly selected PCR-confirmed COVID-19-positive patients from October 1, 2020, to January 30, 2021. All the patient records were taken from the Public Health Reference Lab, Peshawar, Pakistan. Telephonic interviews were conducted for data collection using a structured questionnaire. SPSS version 24.0 was used for data analysis. Regression analysis was done to determine the association between dependent variables (oral manifestations) and independent data like age, gender, and systemic features. RESULTS: Among 117 participants, 56% experienced oral manifestations associated with COVID-19. Taste alterations were experienced by 49.6% (n = 58), xerostomia 34.2% (n = 40), inflammation of the oral cavity 14.5% (n = 17), oral ulcers 10.3% (n = 12), angular cheilitis 6% (n = 7), white lesions 0.9% (n = 1), and bleeding gums 0.9% (n = 1). The average duration for symptoms to last was 10 days. General clinical manifestation of COVID-19 include fever, cough, sore throat, nasal congestion, body aches, diarrhea, difficulty breathing, etc. Oral manifestations had a significant association with gender (p

Published
2023
Full Text
View/download PDF

35. Nanozyme-based sensing of dopamine using cobalt-doped hydroxyapatite nanocomposite from waste bones

Author

Umar Nishan, Nighat Jabeen, Amir Badshah, Nawshad Muhammad, Mohibullah Shah, Irfan Ullah, Saifullah Afridi, Jibran Iqbal, Muhammad Asad, Riaz Ullah, Essam A. Ali, Sarfraz Ahmed, and Suvash Chandra Ojha

Subjects
cobalt-doped hydroxyapatite, hydrogen peroxide, peroxidase mimic, dopamine, colorimetric sensor, Biotechnology, TP248.13-248.65
Abstract

Dopamine is one of the most important neurotransmitters and plays a crucial role in various neurological, renal, and cardiovascular systems. However, the abnormal levels of dopamine mainly point to Parkinson’s, Alzheimer’s, cardiovascular diseases, etc. Hydroxyapatite (HAp), owing to its catalytic nature, nanoporous structure, easy synthesis, and biocompatibility, is a promising matrix material. These characteristics make HAp a material of choice for doping metals such as cobalt. The synthesized cobalt-doped hydroxyapatite (Co-HAp) was used as a colorimetric sensing platform for dopamine. The successful synthesis of the platform was confirmed by characterization with FTIR, SEM, EDX, XRD, TGA, etc. The platform demonstrated intrinsic peroxidase-like activity in the presence of H2O2, resulting in the oxidation of 3,3′,5,5′-tetramethylbenzidine (TMB). The proposed sensor detected dopamine in a linear range of 0.9–35 μM, a limit of detection of 0.51 µM, limit of quantification of 1.7 µM, and an R2 of 0.993. The optimization of the proposed sensor was done with different parameters, such as the amount of mimic enzyme, H2O2, pH, TMB concentration, and time. The proposed sensor showed the best response at 5 mg of the mimic enzyme, pH 5, 12 mM TMB, and 8 mM H2O2, with a short response time of only 2 min. The fabricated platform was successfully applied to detect dopamine in physiological solutions.

Published
2024
Full Text
View/download PDF

36. A comprehensive empirical analysis on cross-domain semantic enrichment for detection of depressive language

Author

Farruque, Nawshad, Goebel, Randy, and Zaiane, Osmar

Subjects
Computer Science - Computation and Language, Computer Science - Artificial Intelligence, Computer Science - Machine Learning
Abstract

We analyze the process of creating word embedding feature representations designed for a learning task when annotated data is scarce, for example, in depressive language detection from Tweets. We start with a rich word embedding pre-trained from a large general dataset, which is then augmented with embeddings learned from a much smaller and more specific domain dataset through a simple non-linear mapping mechanism. We also experimented with several other more sophisticated methods of such mapping including, several auto-encoder based and custom loss-function based methods that learn embedding representations through gradually learning to be close to the words of similar semantics and distant to dissimilar semantics. Our strengthened representations better capture the semantics of the depression domain, as it combines the semantics learned from the specific domain coupled with word coverage from the general language. We also present a comparative performance analyses of our word embedding representations with a simple bag-of-words model, well known sentiment and psycholinguistic lexicons, and a general pre-trained word embedding. When used as feature representations for several different machine learning methods, including deep learning models in a depressive Tweets identification task, we show that our augmented word embedding representations achieve a significantly better F1 score than the others, specially when applied to a high quality dataset. Also, we present several data ablation tests which confirm the efficacy of our augmentation techniques., Comment: This is an extension over ECML-PKDD, 2019 paper "Augmenting Semantic Representation of Depressive Language: from Forums to Microblogs", with more embedding mapping/augmentation methods and data ablation tests. These experiments were done in the year 2019

Published
2021

37. STEP-EZ: Syntax Tree guided semantic ExPlanation for Explainable Zero-shot modeling of clinical depression symptoms from text

Author

Farruque, Nawshad, Goebel, Randy, Zaiane, Osmar, and Sivapalan, Sudhakar

Subjects
Computer Science - Computation and Language, Computer Science - Artificial Intelligence, Computer Science - Information Retrieval, Computer Science - Machine Learning
Abstract

We focus on exploring various approaches of Zero-Shot Learning (ZSL) and their explainability for a challenging yet important supervised learning task notorious for training data scarcity, i.e. Depression Symptoms Detection (DSD) from text. We start with a comprehensive synthesis of different components of our ZSL modeling and analysis of our ground truth samples and Depression symptom clues curation process with the help of a practicing clinician. We next analyze the accuracy of various state-of-the-art ZSL models and their potential enhancements for our task. Further, we sketch a framework for the use of ZSL for hierarchical text-based explanation mechanism, which we call, Syntax Tree-Guided Semantic Explanation (STEP). Finally, we summarize experiments from which we conclude that we can use ZSL models and achieve reasonable accuracy and explainability, measured by a proposed Explainability Index (EI). This work is, to our knowledge, the first work to exhaustively explore the efficacy of ZSL models for DSD task, both in terms of accuracy and explainability., Comment: Fixed an algorithm

Published
2021

44. Post-mortem investigation of deaths due to pneumonia in children aged 1–59 months in sub-Saharan Africa and South Asia from 2016 to 2022: an observational study

Author

Adam, Yasmin, Agaya, Janet, Ahmed, A.S.M. Nawshad Uddin, Ahmed, Dilruba, Alemu, Addisu, Ali, Solomon, Ameh, Soter, Aol, George, Argeseanu, Solveig, Ariuman, Farida, Balogun, Oluseyi, Bari, Sanwarul, Basket, Margaret, Begum, Ferdousi, Bhandari, Manu, Blevins, John, Bunn, James, Bursuc, Courtney, Cain, Carrie Jo, Chawana, Richard, Chawla, Kiranpreet, Chukwuegbo, Cornell, Diarra, Kounandji, Diarra, Tiéman, Diaz, Maureen, Duduyemi, Babatunde, Fairchild, Karen D., Flora, Meerjady Sabrina, Fritz, Ashleigh, Garel, Mischka, Gaume, Brigitte, Gizaw, Mahlet Abayneh, Govender, Nelesh P., Greene, Carol L., Gure, Tadesse, Halu, Binyam, Hoque, Mahbubul, Hwinya, Cleopas, Ibrahim, Alexander M., Igunza, Kitiezo Aggrey, Islam, Ferdousi, Ita, Okokon, Jambai, Amara, Johnson, J. Kristie, Juma, Jane, Kaluma, Erick, Kamal, Mohammed, Kaykay, Osman, Kenneh, Sartie, Khagayi, Sammy, Koka, Rima, Kone, Diakaridia, Koplan, Jeffrey P., Kourouma, Nana, Kowuor, Dickens, LaHatte, Kristin, Lala, Sanjay G., Lee, Kyu Han, Liu, Lucy, Lombaard, Hennie, Maixenchs, Maria, Manhique, Zara, Mannah, Margaret, Martines, Roosecelis, Mash, Ronald, Mehta, Ashka, Menéndez, Clara, Misore, Thomas, Mocumbi, Sibone, Moseray, Andrew, Moses, Francis, Muga, Christopher, Munguambe, Khátia, Myburgh, Nellie, Nair, Shailesh, Ndagurwa, Pedzisai, Nhacolo, Ariel, Nhampossa, Tacilta, Nwajiobi, Princewill, Ochola, Christine, Oliech, Richard, Oluoch, Bernard, Onwuchekwa, Uma U., Onyango, Peter Nyamthimba, Orlien, Stian MS, Otieno, Peter, Oundo, Joseph, Owuor, Harun, Parveen, Shahana, Petersen, Karen, Pratt, Samuel, Rahman, Mahbubur, Rahman, Mohammad Mosiur, Rahman, Mustafizur, Raymer, Sarah, Ritter, Jana, Salzberg, Navit T., Samura, Solomon, Sannoh, Sulaiman, Sanogo, Doh, Seppeh, Martin, Sesay, Tom, Sesay, Joseph Kamanda, Shirin, Tahmina, Sissoko, Seydou, Smart, Francis, Sorour, Gillian, Squire, James, Swaray-Deen, Alim, Swart, Peter J., Tarawally, Fatmata Bintu, Tasnim, Saria, Temesgen, Fikremelekot, Tennant, Sharon M., Traore, Cheick Bougadari, Traore, Awa, Velaphi, Sithembiso, Vyas, Kurt, Wadhwa, Ashutosh, Wadula, Jeannette, Waller, Jessica, Wanga, Valentine, Warang, Shamta, Were, Joyce Akinyi, Wilson, Tais, Winchell, Jonas, Wise, Amy, Witherbee, Jakob, Yeshi, Melisachew Mulatu, Zaman, K., Mahtab, Sana, Blau, Dianna M, Madewell, Zachary J, Ogbuanu, Ikechukwu, Ojulong, Julius, Lako, Sandra, Legesse, Hailemariam, Bangura, Joseph S, Bassat, Quique, Mandomando, Inacio, Xerinda, Elisio, Fernandes, Fabiola, Varo, Rosauro, Sow, Samba O, Kotloff, Karen L, Tapia, Milagritos D, Keita, Adama Mamby, Sidibe, Diakaridia, Onyango, Dickens, Akelo, Victor, Gethi, Dickson, Verani, Jennifer R, Revathi, Gunturu, Scott, J Anthony G, Assefa, Nega, Madrid, Lola, Bizuayehu, Hiwot, Tirfe, Tseyon Tesfaye, El Arifeen, Shams, Gurley, Emily S, Islam, Kazi Munisul, Alam, Muntasir, Zahid Hossain, Mohammad, Dangor, Ziyaad, Baillie, Vicky L, Hale, Martin, Mutevedzi, Portia, Breiman, Robert F, Whitney, Cynthia G, and Madhi, Shabir A

Published
2024
Full Text
View/download PDF

46. Basic and Depression Specific Emotion Identification in Tweets: Multi-label Classification Experiments

Author

Farruque, Nawshad, Huang, Chenyang, Zaiane, Osmar, and Goebel, Randy

Subjects
Computer Science - Machine Learning, Computer Science - Artificial Intelligence
Abstract

In this paper, we present empirical analysis on basic and depression specific multi-emotion mining in Tweets with the help of state of the art multi-label classifiers. We choose our basic emotions from a hybrid emotion model consisting of the common emotions from four highly regarded psychological models of emotions. Moreover, we augment that emotion model with new emotion categories because of their importance in the analysis of depression. Most of those additional emotions have not been used in previous emotion mining research. Our experimental analyses show that a cost sensitive RankSVM algorithm and a Deep Learning model are both robust, measured by both Macro F-measures and Micro F-measures. This suggests that these algorithms are superior in addressing the widely known data imbalance problem in multi-label learning. Moreover, our application of Deep Learning performs the best, giving it an edge in modeling deep semantic features of our extended emotional categories., Comment: Accepted at CICLing, 2019

Published
2021

Catalog

Books, media, physical & digital resources
See catalog results