65 skills found · Page 1 of 3
microsoft / RASAn open-source implementation of Regional Adaptive Sampling (RAS), a novel diffusion model sampling strategy that introduces regional variability in sampling steps
winddori2002 / DEX TTSDEX-TTS: Diffusion-based EXpressive TTS with Style Modeling on Time Variability
Okes2024 / Predicting Groundwater Iron Concentration From Borehole DataThis study aimed to model and predict iron concentrations in groundwater within Yenagoa, Bayelsa State, Nigeria, using machine learning techniques. It focused on evaluating spatial variability and determining the most influential predictors to support groundwater quality management.
realmichaelye / Stress Prediction Using HRVUsing the SWELL dataset from Kaggle, we've built 2 machine learning models to predict whether or not a person is under stress using Heart Rate Variability(HRV) which can be collected from modern wearables such as fitbit devices and apple watches.
Geraldine-Winston / Building Energy Consumption Forecasting Under Climate Variability Using LSTM.This project forecasts building energy consumption using LSTM models, incorporating climate variables like temperature and humidity. It enhances prediction accuracy under seasonal and daily variability for improved energy management and planning.
lbugnon / EmoHRMethods for continuous emotion recognition based on heart rate variability signal. Affect recognition, emotion, HRV, ELM, sSOM, dimensional model of affects
yousseftfifha / Groundwater Management Under Climate ChangeThis project aims to study the impact of climate change on groundwater level in Mornag plain in Tunisia. Indeed, in the last few decades, aquifers all over the world have experienced notable water level variability due to the spatiotemporal variability of rainfall and temperature. Therefore, for a reliable groundwater management under climate change context, it is mandatory to analyze and estimate its level variability. In this study, we focus on the plain of Mornag, located in the southeast of Tunisia, since it represents 33% of the national agricultural production. From this plain, we have collected historical piezometric and pluviometric data covering the period 2005-2017. Knowing the pluviometric data, our goal is to predict the piezometric one. This issue has been already studied using classical numerical groundwater modeling such as Modflow and Feflow. Despite unsatisfactory results, these techniques are data and time consuming. To overcome all these drawbacks, we propose to use two Artificial Intelligence (AI) approaches: the Extreme Gradient Boosting (XGBoost) approach, that has shown great performances in literature, and the well used one in our context which involves the use of Long-Short Term Memory (LSTM) Neural Network. For better results, we have added supplementary features to our dataset such as the cluster zone (zones with same characteristics) and the Standardized Precipitation Index (SPI) which can identify drought at different time scales. Both approaches have been executed entirely on GPU for time acceleration. Compared with traditional existing methods, they both have shown a high level of accuracy which confirms their adequacy for groundwater level forecasting. The proposed prediction models will be used for evaluating the repercussions of climate change on groundwater levels under the different scenarios RCP 4.5 and RCP 8.5 for the period of 2017-2090. It will be evaluated for three future periods: 2017-2040 (short term), 2041-2065 (medium term) and 2066-2090 (long term). The analysis of the future results using AI will be considered as a new Decision Support System used to optimize the management of our limited resources in order to satisfy the needs of the population in terms of drinking water and agriculture production.
NCAR / CVDP NclThe Climate Variability Diagnostics Package (CVDP) developed by NCAR's Climate Analysis Section is an analysis tool that documents the major modes of climate variability in models and observations.
danfenghong / ALMM TIPDanfeng Hong, Naoto Yokoya, Jocelyn Chanussot, Xiaoxiang Zhu. An Augmented Linear Mixing Model to Address Spectral Variability for Hyperspectral Unmixing, IEEE TIP, 2019.
guglielmocamporese / Learning Invariances In Speech RecognitionIn this work I investigate the speech command task developing and analyzing deep learning models. The state of the art technology uses convolutional neural networks (CNN) because of their intrinsic nature of learning correlated represen- tations as is the speech. In particular I develop different CNNs trained on the Google Speech Command Dataset and tested on different scenarios. A main problem on speech recognition consists in the differences on pronunciations of words among different people: one way of building an invariant model to variability is to augment the dataset perturbing the input. In this work I study two kind of augmentations: the Vocal Tract Length Perturbation (VTLP) and the Synchronous Overlap and Add (SOLA) that locally perturb the input in frequency and time respectively. The models trained on augmented data outperforms in accuracy, precision and recall all the models trained on the normal dataset. Also the design of CNNs has impact on learning invariances: the inception CNN architecture in fact helps on learning features that are invariant to speech variability using different kind of kernel sizes for convolution. Intuitively this is because of the implicit capability of the model on detecting different speech pattern lengths in the audio feature.
ananya2001gupta / Bitcoin Price Prediction Using AI ML.Identify the software project, create business case, arrive at a problem statement. REQUIREMENT: Window XP, Internet, MS Office, etc. Problem Description: - 1. Introduction of AI and Machine Learning: - Artificial Intelligence applies machine learning, deep learning and other techniques to solve actual problems. Artificial intelligence (AI) brings the genuine human-to-machine interaction. Simply, Machine Learning is the algorithm that give computers the ability to learn from data and then make decisions and predictions, AI refers to idea where machines can execute tasks smartly. It is a faster process in learning the risk factors, and profitable opportunities. They have a feature of learning from their mistakes and experiences. When Machine learning is combined with Artificial Intelligence, it can be a large field to gather an immense amount of information and then rectify the errors and learn from further experiences, developing in a smarter, faster and accuracy handling technique. The main difference between Machine Learning and Artificial Intelligence is , If it is written in python then it is probably machine learning, If it is written in power point then it is artificial intelligence. As there are many existing projects that are implemented using AI and Machine Learning , And one of the project i.e., Bitcoin Price Prediction :- Bitcoin (₿ ) (founder - Satoshi Nakamoto , Ledger start: 3 January 2009 ) is a digital currency, a type of electronic money. It is decentralized advanced cash without a national bank or single chairman that can be sent from client to client on the shared Bitcoin arrange without middle people's requirement. Machine learning models can likely give us the insight we need to learn about the future of Cryptocurrency. It will not tell us the future but it might tell us the general trend and direction to expect the prices to move. These machine learning models predict the future of Bitcoin by coding them out in Python. Machine learning and AI-assisted trading have attracted growing interest for the past few years. this approach is to test the hypothesis that the inefficiency of the cryptocurrency market can be exploited to generate abnormal profits. the application of machine learning algorithms to the cryptocurrency market has been limited so far to the analysis of Bitcoin prices, using random forests , Bayesian neural network , long short-term memory neural network , and other algorithms. 2. Applications/Scope of AI and Machine Learning :- a) Sentiment Analysis :- It is the classification of subjective opinions or emotions (positive, negative, and neutral) within text data using natural language processing. b) It is Characterized as a use of computerized reasoning where accessible data is utilized through calculations to process or help the handling of factual information. BITCOIN PRICE PREDICTION USING AI AND MACHINE LEARNING: - The main aim of this is to find the actual Bitcoin price in US dollars can be predicted. The chance to make a model equipped for anticipating digital currencies fundamentally Bitcoin. # It works the prediction by taking the coinMarkup cap. # CoinMarketCap provides with historical data for Bitcoin price changes, keep a record of all the transactions by recording the amount of coins in circulation and the volume of coins traded in the last 24-hours. # Quandl is used to filter the dataset by using the MAT Lab properties. 3. Problem statement: - Some AI and Machine Learning problem statements are: - a) Data Privacy and Security: Once a company has dug up the data, privacy and security is eye-catching aspect that needs to be taken care of. b) Data Scarcity: The data is a very important aspect of AI, and labeled data is used to train machines to learn and make predictions. c) Data acquisition: In the process of machine learning, a large amount of data is used in the process of training and learning. d) High error susceptibility: In the process of artificial intelligence and machine learning, the high amount of data is used. Some problem statements of Bitcoin Price Prediction using AI and Machine Learning: - a) Experimental Phase Risk: It is less experimental than other counterparts. In addition, relative to traditional assets, its level can be assessed as high because this asset is not intended for conservative investors. b) Technology Risks: There is a technological risk to other cryptocurrencies in the form of the potential appearance of a more advanced cryptocurrency. Investors may simply not notice the moment when their virtual assets lose their real value. c) Price Variability: The variability of the value of cryptocurrency are the large volumes of exchange trading, the integration of Bitcoin with various companies, legislative initiatives of regulatory bodies and many other, sometimes disregarded phenomena. d) Consumer Protection: The property of the irreversibility of transactions in itself has little effect on the risks of investing in Bitcoin as an asset. e) Price Fluctuation Prediction: Since many investors care more about whether the sudden rise or fall is worth following. Bitcoin price often fluctuates by more than 10% (or even more than 30%) at some times. f) Lacks Government Regulation: Regulators in traditional financial markets are basically missing in the field of cryptocurrencies. For instance, fake news frequently affects the decisions of individual investors. g) It is difficult to use large interval data (e.g., day-level, and month-level data) . h) The change time of mining difficulties is much longer. Moreover, do not consider the news information since it is hard to determine the authenticity of a news or predict the occurrence of emergencies.
byuflowlab / Iea37 Wflo CasestudiesTwo case studies designed to study variability in optimization approach and wake models for wind farm layout optimization. In accordance with IEA Task 37.
AlexTISYoung / HlmmLibrary and command line scripts for fitting heteroskedastic linear mixed models to genetic data. Can be used to perform GWAS for genetic effects on phenotypic variability.
Orang-utan / CNN Sleep Stage Estimation😴 Classifying patient sleep stage (i.e. REM sleep, shallow, etc.) based on Heart Rate Variability data; model is based on time-domain Convolutional Neural Network, implemented in Tensorflow, Python.
shuaikaishi / PGMSUShuaikai Shi, Min Zhao, Lijun Zhang, Yoann Altmann and Jie Chen, "Probabilistic Generative Model for Hyperspectral Unmixing Accounting for Endmember Variability," in IEEE Transactions on Geoscience and Remote Sensing, vol. 60, pp. 1-15, 2022, Art no. 5516915, doi: 10.1109/TGRS.2021.3121799.
Universal-Variability-Language / Uvl ParserUVL (Universal Variability Language) is a concise and extensible language for modeling variability in software product lines. This repository is holding the grammar definition.
zmlabe / ExtremeEventsUsing ANN's to reveal changes in extreme events and internal variability in climate models
ivky03 / Wind Power Forecasting Using Ensemble LearningAn accurate and reliable wind power forecasting model that can handle the variability and uncertainty of the wind resource. An ensemble model which includes the Transformer, LSTM and Gradient Boosting Decision Tree models.
VariantSync / VatrasAgda Library to Study the Expressive Power of Languages for Static Variability
dirge1 / FBM ADTcode of the paper "Reliability modeling and statistical analysis of accelerated degradation process with memory effects and unit-to-unit variability" fractional Brownian motion, random effects, wiener process, stochastic, expectation maximization, EM algorithm, parameter estimation, FBM, uncertainty quantification, assessment, non-Markovian
