https://github.com/vanessaaleung/support-ticket-nlp

Support Ticket Classification and Key Phrases Extraction
https://github.com/vanessaaleung/support-ticket-nlp

keras machine-learning python text-classification ticket-classification

Last synced: 2 days ago
JSON representation

Support Ticket Classification and Key Phrases Extraction

• Host: GitHub
• URL: https://github.com/vanessaaleung/support-ticket-nlp
• Owner: vanessaaleung
• Created: 2020-05-20T02:04:34.000Z (over 4 years ago)
• Default Branch: master
• Last Pushed: 2020-06-24T01:14:43.000Z (over 4 years ago)
• Last Synced: 2024-11-10T22:14:14.254Z (2 months ago)
• Topics: keras, machine-learning, python, text-classification, ticket-classification
• Language: Jupyter Notebook
• Homepage:
• Size: 4.94 MB
• Stars: 4
• Watchers: 1
• Forks: 1
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • Support: Support_Ticket_NLP_Case.ipynb

Awesome Lists containing this project

README

        
# Support Ticket NLP

Support Ticket Classification and Key Phrases Extraction

- Identify the main issues in the ticket description

- Extract the key phrases in the ticket description



----------------

## Data

Support Ticket Classification



----------------

## Tasks



  
Topic Modeling with LDA model

    


      


      Preprocessing

        


          
Divide text to tokens

          
Remove stopwords, punctuations

          
Lemmatization

        

      

      
Compute coherence values to find the optimal number of topics

      

      
Build the LDA model

      

      
Utilize pyLDAvis to visualize the topics

    

  

  
Key Phrases Extraction with pytextrank (combining spaCy and networkx)

    


      
Construct a graph, sentence by sentence, based on the spaCy part-of-speech tags tags

      
Use matplotlib to visualize the lemma graph

      

      
Use PageRank – which is approximately eigenvalue centrality – to calculate ranks for each of the nodes in the lemma graph

        


          
$a_{v,t}=1$ if vertex $v$ is linked to vertex $t$, and $a_{v,t}=0$ otherwise

          
$M(v)$ is a set of the neighbors of $v$ and $\lambda$ is a constant

        

      

      

      
Collect the top-ranked phrases from the lemma graph based on the noun chunks

      
Find a minimum span for each phrase based on combinations of lemmas

        


        permission 1 0.17555037929471423

        requisitions 1 0.1742458175386728

        recruiter 1 0.1416381454134179

        

      

    

  



  

----------------

## Terminologies

### Topic Coherence

Scores a single topic by measuring the degree of semantic similarity between high scoring words in the topic


### Latent Dirichlet Allocation (LDA)

Given the # documents, # words, and # topics, output:

1. distribution of words for each topic K

2. distribution of topics for each document i