Extracting numeric token for several entities in order using Spacy

, ,
1

Hi All,

I am new to information extraction and especially to Spacy so my apologies in advance if the question is too naive.

I have the following text:

In the last year, Medical AI firms had raised even more, with AICare and DeepScan, raising £2million and £5million, respectively.

The NER model shows it as follows:

Screen Shot 2020-10-01 at 21.32.58
Screen Shot 2020-10-01 at 21.32.581734×132 61 KB

I want to extract information in the following format:

Screen Shot 2020-10-01 at 21.34.11
Screen Shot 2020-10-01 at 21.34.11776×132 30.8 KB

The starter code is

text = "In last year, Medical AI startups had won even more, with AICare and DeepScan, raising $2million and $5million, respectively."

def get_nlp():
 
    nlp = spacy.load("en_core_web_sm")

    ruler = EntityRuler(nlp, overwrite_ents=True)

    patterns = [{"label": "RESEARCHSTRAND", "pattern":[{"lower": "medical"}, {"lower": "ai"}]},
                {"label": "COMPANY", "pattern":[{"lower": "aicare"}]},
                {"label": "COMPANY", "pattern":[{"lower": "deepscan"}]},    
               ]

    ruler.add_patterns(patterns)
    nlp.add_pipe(ruler, before="ner")

    return nlp(raw_text)

doc = get_nlp(text)
displacy.render(doc, jupyter=True, style='ent')

Any idea on how to achieve this task. I will greatly appreciate your support.

Many Thanks and

Kind Regards,
Muhhas