Usage Notebooks

Contents:

• Preprocessing Demo
  • Text to clean and numerize
  • Pipeline creation
  • Coherent text example with Autocorrect and ContextualSpellCheck spelling correction
  • PySpellChecker
  • ContextualSpellCheck: Setup
• Abbreviation Handler Demo
  • AbbrExpander class
  • Abbreviation class
• Pipelines
  • Internal Developed Functions
  • Reset NLP Pipeline
  • Example
• Similarity Demo
  • Setup the path and load DACKAR modules for similarity analysis
  • Example
  • Compute sentences similarity without disambiguation
  • Compute sentences similarity with disambiguation
  • Convert sentences into synsets list, and then compute similarity
• NLP Alias Demo
• ER Schema
• MBSE
• Rule-based NLP
  • 1. Set up the path, so that the NLP modules can be found
  • 2. Load Spacy module
  • 3. Load other modules
  • 4. Import NLP modules
  • 5. Set up logging
  • 6. Read and process entities
  • 7. Read and process causal keywords
  • 8. Create Rule-based matcher with entity list and causal entity list
  • 9. Read input text file, or users can provide a raw string
  • 10. Process raw string data using matcher
  • 11. Access processed information from matcher
• Workflow Demo
  • Set Paths and Loading Required Modules
  • Initialize variables
  • Load entity list and causal list or provide directly
  • Generate patterns that can be used in NER
  • Create rule-based matcher with entity list and causal entity list
  • Read raw text data and preprocess it
  • Correct the doc
  • Process text using Rule Based Matcher
• Work Order Processing
  • Setup path and load DACKAR modules
  • Generate entities patterns and process text using WorkOrderProcessing class
  • Processing work order accumulatively
  • Accessing attributes of entities
• Anomaly Detection
  • Anomaly types and matrix profile can convert different types fo anomalies into outliers
  • Set up paths and load matrix profile module
  • Calculate the matrix profiles for NY taxi data
  • Calculate matrix profile for steam generator data
  • Testing ‘approx’ method to compute matrix profile
  • Enable Streaming, use ‘evaluate’ function for streaming data
  • Test different data structure
  • Test Multi-Dimensional Anomaly Detection: Identify Best K out of N Anomalies
• Knowledge Graph Demo
  • Demo of knowledge graph construction
  • Demo of time series anomaly detection
  • Multi-dimensional anomaly detection: identify best K out of N anomalies
  • Technical Language Processing or Natural Language Processing