`relevanceai.operations_new.cooccurrence_network.transform`#

Build a co-occurrence network based on the documents in your dataset

Module Contents#

relevanceai.operations_new.cooccurrence_network.transform.preprocess(data, stopwords_list=[])#

class relevanceai.operations_new.cooccurrence_network.transform.WordDictionary(docs)#

get_df_table(self)#

get_ids(self, doc_id)#

get_docs(self, word: str)#

get_word(self, id)#

get_id(self, word: str)#

update_df_table(self, word: str)#

class relevanceai.operations_new.cooccurrence_network.transform.CoOccurNetTransform(max_number_of_clusters=15, min_number_of_clusters=3, number_of_concepts=100, **kwargs)#

To write your own operation, you need to add: - name - transform

find_max_vertex(self, visited, weights)#

maximum_spanning_tree(self, graph)#

concurrence_matrix(self, top_ids, texts, word_dict)#

get_clusters_labels(self, mat)#

transform(self, documents, text_field='content', stopwords_list=[], center_word=None)#: abstractmethod for transform

property name(self)#: abstractproperty for name

relevanceai.operations_new.cooccurrence_network.transform#

Module Contents#

`relevanceai.operations_new.cooccurrence_network.transform`#