D1.Project Definition
Purpose:
Analyse large amounts of texts and data, e.g. quality reports, customer reviews/complaints to identify the most severe issues. A new project is defined to address those issues.
Procedure:
1) Various text mining techniques can help analyse large amounts of texts and data, e.g. quality reports, customer reviews/complaints.
2a) Use a topic detection algorithm, such as LDA (Latent Dirichlet Allocation), to identify relevant themes in text documents.
2b)Texts can be converted into document vectors so machine learning algorithms can be applied to text data.
Use clustering algorithms like DB scan using Levenshtein or Cosine distance to group salient topics.
3) Analyse identified themes/topics, i.e. by sorting or ranking them to define a new project.
