• Developed deep learning and generative AI methods for emotion recognition in multimedia data using multimodal fusion techniques
• Utilized audio-based deep learning models for audio signal processing and similarity detection
• Implemented deep learning methods for speech recognition and emotion detection in social media videos
• Fine-tuned transformer models for detecting and calculating toxicity scores with NLP techniques
• Analyzed the spread of toxicity on social media using generative AI models and epidemiological modeling
• Develop a novel epidemiological model to study hate speech
• Developed methods for duplicate detection in NLP using transformers and LLM-based models
H
Hushly
Data scientist
Cupertino, CA, United States, 95015
June 2022 - July 2022
P
Pars Net company
Data Analyst
January 2016 - December 2017
Skills
AlgorithmsArtificial IntelligenceArtificial Neural NetworksAudio Signal ProcessingBusiness Process ModelingCase StudiesCluster AnalysisComputer ProgrammingCOVID-19 TestingC++ (Programming Language)Data AnalysisData MiningData ScienceDeep LearningDesktop ComputingElectrical TransformersEpidemiologyEvent ManagementFeature EngineeringForecasting SkillsGPTImage ProcessingInformation TechnologyKaggleKerasLarge Language ModelsMachine LearningMATLABMatplotlibMultimediaNatural Language ProcessingNetwork AnalysisNumPyOncologyPandasPython (Programming Language)PytorchResonanceSalesScikit LearnSignal ProcessingSocial MediaSocial Network AnalysisTensorflowTrustworthinessWeka