Parsing Incoming Applications with NLP
summary
Nikolay Knyazev presents using natural language processing to analyze incoming requests for robotic process automation workflows. The talk covers applying NLP techniques to parse and categorize user or customer requests so RPA bots can route, triage, or handle them automatically.
AI-written summary · about
chapters (25)
- 0:00Welcome
- 5:53Parsing email with NLP
- 6:37My ML experience
- 8:30Email classification
- 10:19Text classification in business
- 12:08No-code ML approaches
- 14:20Text analysis solutions
- 15:47How ML processes text
- 17:13Text normalization
- 20:26Text to features
- 23:43TF-IDF vs neural
- 24:50Transformers revolution
- 25:54Supervised vs unsupervised
- 27:04Stack comparison
- 28:36Model comparison
- 29:41Multiclass classification
- 31:06AutoML pros and cons
- 32:10Hierarchical classification
- 35:07Quick deployment
- 36:15Custom solution costs
- 37:40ML results
- 38:42ML integration
- 39:27Entity extraction
- 41:34Entity extraction challenges
- 42:36Entity extraction process
related talks
55mDemystifying Natural Language Processing across several industry verticals
06mAdvancements in Natural Language Processing (NLP) and Text Analytics
57mNatural Language Processing in .NET
26mNatural Language Processing (NLP) going mainstream
12mUnlocking the Power of NLP: Techniques in K8s-Native Applications
36mCrash Course in Machine Learning for Natural Language Processing
31mNatural Language Processing & Java
20mNatural Language Interfaces for SaaS: Transforming Workflows with ConversationalAI