Seasoned Data Scientist experienced working with large datasets, breaking down information and applying interpretations to complex business concerns. Proficient in distribution, predictive and hypothetical modeling. Bring…
Seasoned Data Scientist experienced working with large datasets, breaking down information and applying interpretations to complex business concerns. Proficient in distribution, predictive and hypothetical modeling. Bringing 4+ years of related experience strengthening company operations.
AI driven Asset Health Care Solution to achieve top asset performance and reliability, utilizing Machine Learning or Deep Learning Models that provides early prediction to reduce failures and avoid unplanned maintenance/ plant shutdowns.
The Model is trained initially on historical data, predicts the results for next one year data for validation model will be predicted with subsequent year unseen test data and if fine required will be applied K-means algorithm is used to find the clusters and the outlier score is calculated using CMGOS algorithm. If baseline approach does not achieve the acceptance criteria, AE-based approach is applied on the asset.
The anomaly predicted by the model will be validated against validation data taken out from historical data. Anomaly feedback was collected and performance of model was monitored. The performance of the model is expected to match the historical performance of the model. After the model achieves an acceptable precision and recall criteria it has pushed to pre prod for live predictions
AI driven Asset Health Care Solution to achieve top asset performance and reliability, utilizing Machine Learning or Deep Learning Models that provides early prediction to reduce failures and avoid unplanned maintenance/ plant shutdowns.
Develop a model initially trained on historical data to predict results using the data from the subsequent one year for validation.
The model will then be tested with unseen data from the subsequent year.
If the results are satisfactory, K-means algorithm will be applied to identify clusters, and outlier scores will be calculated using the CMGOS algorithm.
In cases where the baseline approach fails to meet the acceptance criteria, an AE-based approach will be employed on the asset.
The anomaly predicted by the model will be validated against validation data taken out from historical data.
Anomaly feedback was collected and performance of model was monitored.
The performance of the model is expected to match the historical performance of the model.
After the model achieves an acceptable precision and recall criteria it has pushed to pre prod for live predictions.
Text auto-completion involves extracting, preprocessing, and tokenizing textual data from diverse sources.
This process refines data quality by eliminating special characters, punctuation, and stopwords.
Advanced techniques such as stemming, lemmatization, and word embeddings are applied to transform words into numerical representations suitable for deep learning.
These methods facilitate efficient model training and improve understanding of semantic context, thereby enhancing the model's accuracy and predictive capability.
Developed and trained deep learning models using RNNs or transformers with TensorFlow or PyTorch.
Employed techniques like word embeddings, LSTM layers, attention mechanisms, and dropout regularization to capture sequential patterns and long-range dependencies in text data.
Fine-tuned hyperparameters and evaluated model performance using metrics such as accuracy, and loss on validation datasets for optimal predictive capability.
نظام التوظيف التشغيلي المعتمد على الذكاء الاصطناعي أولاً، الذي يوحّد ويؤتمت دورة التوظيف بالكامل للمؤسسات الحديثة حول العالم.
د.جوب منطقة حرة ذ.م.م. 2026 © جميع الحقوق محفوظة
