Init commit

add_embeddings.py

+#!/usr/bin/env python
+# coding: utf-8
+
+# Imports
+import json
+import spacy
+import numpy as np 
+import sys, getopt
+
+def sentence_embedding(sentences):
+    """ Computes the embeddings of a list of sentences, omitting 
+        stop_words and punctuation symbols """
+    nlp = spacy.load("en_core_web_lg")
+    sentence_embeddings = []
+    
+    for doc in nlp.pipe(sentences, disable=["tagget", "parser", "ner"]):    
+        tokens = list(filter(lambda t: t.has_vector and (not t.is_punct) and (not t.is_stop), doc.doc)) 
+        vectors = list(map(lambda t: t.vector, tokens))
+        embedding = np.average(vectors, axis=0)
+        sentence_embeddings.append(embedding)
+        
+    
+    return np.array(sentence_embeddings)
+
+
+def save_as_json(file_name, data):
+    """ Saves the provided data as a json file """
+    with open(file_name, 'w', encoding='utf8') as fout:
+        json.dump(data , fout, indent=4, sort_keys=True, ensure_ascii=False)
+
+
+def read_json(file_name):
+    """ Reads a json file and returns it content """
+    with open(file_name, 'r', encoding='utf8') as file:
+        data = json.load(file)
+    
+    return data
+
+
+def main(argv):
+    """ MAIN """ 
+    input_file = ''
+    output_file = ''
+
+    try:
+        opts, args = getopt.getopt(argv,"hi:o:",["ifile=","ofile="])
+    except getopt.GetoptError:
+        print('add_embeddings -i <input_file.json> -o <output_file.json>')
+        sys.exit(2)
+    for opt, arg in opts:
+        if opt == '-h':
+            print('add_embeddings -i <input_file.json> -o <output_file.json>')
+            sys.exit()
+        elif opt in ("-i", "--ifile"):
+            input_file = arg
+        elif opt in ("-o", "--ofile"):
+            output_file = arg
+
+    qa_pairs = read_json(input_file)
+    questions = list(map(lambda pair: pair['question'].lower(), qa_pairs))
+    q_embeddings = sentence_embedding(questions)
+
+    # Add embeddings
+    for i in range(len(qa_pairs)):
+        pair = qa_pairs[i]
+        pair['embedding'] = q_embeddings[i].tolist()
+
+    # Save QA pairs with embeddings
+    save_as_json(output_file, qa_pairs)
+
+if __name__ == '__main__':
+    main(sys.argv[1:])
\ No newline at end of file

generate_qa_pairs.py

+#!/usr/bin/env python
+# coding: utf-8
+
+# IMPORTS
+import os 
+import pandas as pd 
+from re import sub
+from json import dump
+from bs4 import BeautifulSoup 
+from urllib import error
+from urllib.request import Request, urlopen
+from functools import reduce
+
+
+# GLOBAL VARIABLES
+DIR_NAME = 'raw_html'
+OUTPUT_ROOT_NAME = "qa_pairs"
+INTREL_ROOT_NAME = "intrel"
+DOCTORATE_ROOT_NAME = "doctorate"
+INTREL_URLS = ['https://relint.uva.es/internacional/english/students/welcome-guide/faq/',
+               ]
+DOCTORATE_URLS = ['https://escueladoctorado.uva.es/export/sites/doctorado/faqs/AAFF/?lang=en', 
+                  'https://escueladoctorado.uva.es/export/sites/doctorado/faqs/admisionYMatricula/?lang=en',
+                  'https://escueladoctorado.uva.es/export/sites/doctorado/faqs/PD/?lang=en',
+                  'https://escueladoctorado.uva.es/export/sites/doctorado/faqs/tesis/?lang=en',
+                  'https://escueladoctorado.uva.es/export/sites/doctorado/faqs/financiacion/?lang=en',
+                  ]
+
+
+
+# FUNCTIONS
+def gen_file_path( root, i):
+    """ Combines a root name and an index into an HTML file name """
+    return "{}/{}{}.html".format(DIR_NAME, root, i)
+
+
+def download_files():
+    """ Downloads html files if they don't exist and saves
+        them in the specified directory.
+    """
+    # Create directory
+    try:
+        os.makedirs(DIR_NAME)
+    except FileExistsError:
+        # Directory already exists
+        pass
+    
+    # Download UVA International Relationships FAQs
+    for i, url in enumerate(INTREL_URLS):
+        file_path = gen_file_path(INTREL_ROOT_NAME, i+1)
+
+        try:
+            fin = open(file_path, 'rb')
+            fin.close()
+        except FileNotFoundError:
+            # File doesn't exist
+            req = Request(url,  headers={'User-Agent': 'Mozilla/5.0'})
+            webContent = urlopen(req).read()
+            fout = open(file_path, 'wb')
+            fout.write(webContent)
+            fout.close() 
+
+    # Download UVA Doctorate School FAQs
+    for i, url in enumerate(DOCTORATE_URLS):
+        file_path = gen_file_path(DOCTORATE_ROOT_NAME, i+1)
+
+        try:
+            fin = open(file_path, 'rb')
+            fin.close()
+        except FileNotFoundError:
+            # File doesn't exist
+            req = Request(url,  headers={'User-Agent': 'Mozilla/5.0'})
+            webContent = urlopen(req).read()
+            fout = open(file_path, 'wb')
+            fout.write(webContent)
+            fout.close() 
+
+
+def process_a_tags(content):
+    """ Processes each <a> tag extracting its 'href' attribute and adding 
+        it between parenthesis after the text. It considers the existence
+        of a <base> tag.
+    """
+    base = content.find('base')
+  
+    for a in content.find_all('a'):
+        if a.previousSibling:
+            href = a['href']
+            text = a.get_text()
+            if base != None:
+                url = urljoin(base['href'], href)
+            else:
+                url = href            
+            a.previousSibling.replaceWith(a.previousSibling + " {} ({}) ".format(text, url)) 
+            a.extract() 
+
+def extract_qa_pairs_intrel(page):
+    """ Extracts all the QA pairs of a intrelX.html file and 
+        creates a list of dicts with 'answer', 'question' and 'section' values
+    """
+    qa_pairs = [] 
+    content = BeautifulSoup(page, "html.parser")
+    process_a_tags(content)
+    
+    sec_titles = list(map(lambda x : x.get_text(), 
+                            content.find_all('div', class_='elementor-clearfix')))
+    sec_contents = content.find_all('div', {'data-accordion-type':'accordion'})   
+
+    for sec in range(len(sec_titles)):
+        q_is = sec_contents[sec].find_all('i', class_='fa-accordion-icon')
+        q_spans = reduce(lambda a, b: a+b, list(map(lambda x : x.find_parents('span'), q_is)))
+        sec_questions = list(map(lambda x : x.get_text(), q_spans))
+
+        sec_answers = list(map(lambda x : x.get_text(), 
+                            sec_contents[sec].find_all('div', class_=['eael-accordion-content', 'clearfix'])))
+
+        for p in range(len(sec_questions)):
+            qa_pairs.append({'question': sec_questions[p], 
+                            'answer': sec_answers[p], 
+                            'page': 'relint',
+                            'section': sec_titles[sec]
+                            })
+    
+    return qa_pairs
+
+
+def extract_qa_pairs_doctorate(page):
+    """ Extracts all the QA pairs of doctorateX.html file and 
+        creates a list of dicts with 'answer', 'question' and 'section' values
+    """
+    qa_pairs = []        
+    content = BeautifulSoup(page, "html.parser")
+    
+    sec_title = content.find_all('div', class_='headline')[1].get_text()\
+                        .replace('Frequently asked questions about ', '')\
+                        .replace('Frequently Asked Questions about ', '')
+    
+    sec_content = content.find_all('div', class_='panel-group acc-v2')[1]  
+
+    sec_questions = list(map(lambda x: sub('\n[\t]+[\s]+', '', x.get_text()),
+                            sec_content.find_all('a', class_='accordion-toggle')))
+
+    process_a_tags(sec_content)
+
+    sec_answers = list(map(lambda x: x.get_text().replace('\n', ''),
+                           sec_content.find_all('div', class_='panel-body')))
+    
+    for p in range(len(sec_questions)):
+        qa_pairs.append({'question': sec_questions[p], 
+                        'answer': sec_answers[p],
+                        'page': 'doctorate', 
+                        'section': sec_title,
+                         })
+    return qa_pairs
+
+
+def extract_qa_pairs():
+    """ Extracts all the QA pairs and creates a list of dicts with 
+        'answer', 'question' and 'section' values
+    """
+    qa_pairs = [] 
+    
+    for i in range(len(INTREL_URLS)):
+        file_path = gen_file_path(INTREL_ROOT_NAME, i+1)
+        with open(file_path, "r") as fin:
+            page = fin.read()            
+        qa_pairs.append(extract_qa_pairs_intrel(page))
+    
+    for i in range(len(DOCTORATE_URLS)):
+        file_path = gen_file_path(DOCTORATE_ROOT_NAME, i+1)
+        with open(file_path, "r") as fin:
+            page = fin.read()            
+        qa_pairs.append(extract_qa_pairs_doctorate(page))
+
+    return list(reduce(lambda x,y: x+y, qa_pairs)) 
+
+
+def save_as_json(file_name, data):
+    """ Saves the provided data as a json file """
+    with open('{}.json'.format(file_name), 'w', encoding='utf8') as fout:
+        dump(data , fout, indent=4, sort_keys=True, ensure_ascii=False)
+
+
+def save_as_csv(file_name, data):
+    """ Saves the provided data as a csv file """ 
+    questions = list(map(lambda p: p['question'], data))
+    answers = list(map(lambda p: p['answer'], data))
+    sections = list(map(lambda p: p['section'], data))
+    d = {'Section': sections, 'Question': questions, 'Answer': answers }
+    df = pd.DataFrame(data=d)
+    df.to_csv('{}.csv'.format(file_name), sep='\t', index=False)
+
+
+def main():
+    """ MAIN """
+    download_files()
+    qa_pairs = extract_qa_pairs()
+    save_as_json(OUTPUT_ROOT_NAME, qa_pairs)
+    save_as_csv(OUTPUT_ROOT_NAME, qa_pairs)
+
+
+if __name__ == '__main__':
+    main()
\ No newline at end of file

intrel_qa_pairs.csv

+Section	Question	Answer
+STUDIES AT THE UNIVERSITY OF VALLADOLID	CAN I DO A STUDY EXCHANGE AT THE UNIVERSITY OF VALLADOLID?	You should find out whether there is an exchange agreement between our universities. To do this, you should contact  incoming@uva.es (mailto:incoming@uva.es) If there is no agreement between our universities, you can always come as a  Visiting student (http://relint.uva.es/internacional/english/students/visitors-programme/) .
+STUDIES AT THE UNIVERSITY OF VALLADOLID	WHICH COURSES CAN I TAKE AT THE UNIVERSITY OF VALLADOLID?	You can choose the courses you like from any academic course or academic year. You can also even take courses from various curricula and from different faculties at the same time. However, you must make sure that your coordinating teacher accepts your choices when signing your studies contract or Learning Agreement.Whatever you opt to do, you should take at least one course from the University of Valladolid faculty you have chosen.
+STUDIES AT THE UNIVERSITY OF VALLADOLID	CAN I CHOOSE COURSES FROM ANY TERM OR SEMESTER?	No, you can only choose courses taught during the time you are at our university.If you are here between September and February, you can choose courses taught in the first term or semester.If you are here between February and June, you can choose courses taught in the second term or semester.If you are here from September to June, you can choose any course, including full-year courses.
+STUDIES AT THE UNIVERSITY OF VALLADOLID	CAN I CHOOSE FULL-YEAR COURSES IF I STUDY AT THE UVA FOR A TERM OR A SEMESTER?	No. Full-year courses are taught continuously from September until June. You are not allowed to take them if you are only going to be here during the first or second semester.
+STUDIES AT THE UNIVERSITY OF VALLADOLID	CAN I CHOOSE COURSES FROM VARIOUS FACULTIES AT THE SAME TIME?	Yes, provided that your coordinating teacher agrees to this when signing your studies contract or Learning Agreement. However, you must always take at least one course from the University of Valladolid faculty you have chosen.
+STUDIES AT THE UNIVERSITY OF VALLADOLID	CAN I TAKE COURSES AT VARIOUS CAMPUSES AT THE SAME TIME?	You may only choose courses from the Valladolid and Palencia campuses at the same time, provided that your coordinating teacher agrees to this when signing your studies contract or Learning Agreement. All of the other combinations of campuses are over 100 kilometres away from each other.
+STUDIES AT THE UNIVERSITY OF VALLADOLID	I AM A MASTER’S DEGREE STUDENT. CAN I TAKE BACHELOR’S DEGREE COURSES?	Yes, provided that your coordinating teacher agrees to this when signing your studies contract or Learning Agreement.
+STUDIES AT THE UNIVERSITY OF VALLADOLID	WHERE CAN I SEE THE UVA CURRICULA?	"You can see the various bachelor's degree curricula  here (http://www.uva.es/export/sites/uva/2.docencia/2.01.grados/2.01.02.ofertaformativagrados/2.01.02.01.alfabetica/index.html) , and the master's degree curricula  here (http://www.uva.es/export/sites/uva/2.docencia/2.02.mastersoficiales/2.02.01.ofertaeducativa/2.02.01.01.alfabetica/index.html) . Choose the curriculum you are interested in, and then click on ""Asignaturas"" (courses) to see the courses. "
+STUDIES AT THE UNIVERSITY OF VALLADOLID	WHERE CAN I FIND THE TIMETABLE AND CLASSROOMS FOR MY COURSES?	"Each University of Valladolid faculty has its own webpage where the calendars, timetables and classrooms for the courses are published. You can find the link to your centre’s web  here (http://www.uva.es/export/sites/uva/2.docencia/2.01.grados/2.01.02.ofertaformativagrados/2.01.02.01.alfabetica/index.html) .Choose the curriculum you are interested in, and then click on ""Horarios"" (timetables) to see the times and rooms for your courses."
+STUDIES AT THE UNIVERSITY OF VALLADOLID	IN SOME COURSES IT MENTIONS GROUPS SUCH AS 1 T, 2 A, 1 S. WHAT DOES THIS MEAN? CAN I CHOOSE THE GROUP I WANT?	Some courses are divided into groups because not all of the students can fit into the classroom at the same time. Some courses may have more than one group for the lectures, some for practical lessons, and others for both lectures and practical lessons.Exchange students can choose whichever groups suit them best. In other words, they can choose group 1 for the lectures in one course, and group 2 for another, etc. However, once a student has chosen a particular group for a course, they should always go to the same group for that course.
+STUDIES AT THE UNIVERSITY OF VALLADOLID	WHAT IS THE MINIMUM NUMBER AND THE MAXIMUM NUMBER OF CREDITS I CAN TAKE?	You can take a maximum of 30 ECTS (European Credit Transfer and Accumulation System).We do not set a minimum number of credits. If you need to choose a minimum number of credits you must notify your university.
+STUDIES AT THE UNIVERSITY OF VALLADOLID	WHEN SHOULD I SEND MY LEARNING AGREEMENT OR STUDIES PROGRAMME?	You only need to send it to us if we ask you for it. If we don’t ask you for it, it is because we don’t need it yet. Generally speaking, Erasmus+ students can fill in their Learning Agreement and obtain their coordinating teacher’s signature after they get to the University of Valladolid.
+STUDIES AT THE UNIVERSITY OF VALLADOLID	I NEED TO SIGN MY LEARNING AGREEMENT. WHO DO I SEND IT TO?	You should send it directly to your coordinating teacher at the University of Valladolid.
+STUDIES AT THE UNIVERSITY OF VALLADOLID	MY COORDINATING TEACHER ISN’T ANSWERING MY EMAILS. WHO CAN HELP ME?	If you have any questions related to academic matters or if you need your Learning Agreement signed, you can contact your faculty’s International Relations Coordinator at the University of Valladolid.For any other kinds of question, contact  incoming@uva.es (mailto:incoming@uva.es) 
+STUDIES AT THE UNIVERSITY OF VALLADOLID	I WANT TO DO A FULL BACHELOR’S/MASTER’S/DOCTORAL DEGREE AT THE UNIVERSITY OF VALLADOLID. WHERE CAN I GET INFORMATION?	You can find information on the general website  http://www.uva.es/ (http://www.uva.es/) . The enrolment procedure is the same for Spanish students as for students of other nationalities. If you have any doubts, you should contact the  Seccion de Alumnos (http://directorio.uva.es/detalle.html?id=4001004) .
+THE LANGUAGE OF INSTRUCTION	WHAT LANGUAGE OF INSTRUCTION IS USED AT THE UNIVERSITY OF VALLADOLID?	Approximately 90% of the lessons are in Spanish. A few courses are also in English, and there are courses in other languages as foreign languages.
+THE LANGUAGE OF INSTRUCTION	CAN I STUDY IN ENGLISH AT THE UNIVERSITY OF VALLADOLID?	Some degrees have a lot of courses in English, such as the Bachelor's Degree in English Studies. There are also courses taught in English. You can check which ones  here (http://relint.uva.es/wp-content/uploads/2019/03/English18-19.pdf) .You can also take the International Semesters:https://campusvirtual.uva.es/course/index.php?categoryid=309http://www2.emp.uva.es/index.php/relaciones-internacionales-facultad-de-comercio-de-valladolid/https://campusvirtual.uva.es/course/index.php?categoryid=277http://www.feyts.uva.es/?q=node/2424
+THE LANGUAGE OF INSTRUCTION	WHAT LEVEL OF SPANISH DO I NEED TO STUDY AT THE UVA?	You need a B1 level of Spanish (Common European Framework of Reference for Languages - CEFR) when you get to the University of Valladolid.
+THE LANGUAGE OF INSTRUCTION	I HAVE A GOOD COMMAND OF SPANISH BUT DO NOT HAVE AN OFFICIAL B1 CERTIFICATE. CAN I PRESENT OTHER CERTIFICATES OF SPANISH?	Yes. We only need to be sure that you won’t have any difficulty understanding your teachers, such that we accept non-official certificates.
+THE LANGUAGE OF INSTRUCTION	I SPEAK SPANISH BUT DO NOT HAVE CERTIFICATES. WHAT CAN I DO?	You can send us a report from a Spanish language teacher at your university, or from your International Relations Office, stating that you have sufficient command of Spanish to be able to satisfactorily undertake your stay at our university.Another way to acquire a sufficient level of Spanish is to take an intensive course at a language centre or language school in Spain before you start your academic year.
+WHEN YOU GET TO THE UNIVERSITY OF VALLADOLID	WHAT DO I HAVE TO DO THE FIRST DAY I GET TO THE UVA?	If you are going to study in the city of Valladolid, you should go to the International Relations Office to register that you have arrived and we will give you information. You can find our address and the office opening hours through this link.If you are going to study in Palencia, Segovia, or Soria you should visit your coordinating teacher, who will register your arrival and will give you academic information. You should let your coordinating teacher know in advance by email the date on which you will be arriving. You can find their contact details in your Personal Area.If your coordinating teacher is late in answering you, you can also contact the assistant at the International Relations Office on your campus through the following email addresses:Palencia Campus:  palencia@uva.es (mailto:internacional.palencia@uva.es) Segovia Campus:  segovia@uva.es (mailto:internacional.segovia@uva.es) Soria Campus:  soria@uva.es (mailto:internacional.soria@uva.es) 
+WHEN YOU GET TO THE UNIVERSITY OF VALLADOLID	I STILL DO NOT HAVE MY STUDENT CARD AND I NEED IT. WHAT SHOULD I DO?	You can still have access to all university services and facilities even if you are still waiting for your student card.There are two ways to access: with your enrolment sheet or downloading the UVa app for smartphones, available for  Android (https://play.google.com/store/apps/details?id=com.pocketuniversity.uva)  and  iOS (https://itunes.apple.com/app/id1031505989) .
+EXAMINATIONS AND MARKS/GRADES	WHAT MARK/GRADE WILL I GET IN MY COUNTRY WITH THE EXAMINATIONS I TAKE AT THE UNIVERSITY OF VALLADOLID?	Each country has a different grading system. European countries share a transfer system which you can check out in Wikipedia:  ECTS (https://es.wikipedia.org/wiki/European_Credit_Transfer_and_Accumulation_System) ,  Grading scale (only in English) (https://en.wikipedia.org/wiki/ECTS_grading_scale) .
+EXAMINATIONS AND MARKS/GRADES	WHEN ARE EXAMINATIONS HELD AT THE UNIVERSITY OF VALLADOLID?	First term examinations are held in January, although some teachers may bring theirs forward to late December.Second term examinations are held in June.
+EXAMINATIONS AND MARKS/GRADES	I AM NOT IN SPAIN ON THE DAY OF THE EXAM. CAN I TAKE THE EXAM IN MY HOME COUNTRY?	You should ask your course teacher. They will tell you if there is another way for you to take the exam.
+EXAMINATIONS AND MARKS/GRADES	I DID NOT PASS AN EXAM. WHAT CAN I DO?	In Valladolid, you have the chance to take an exam again if you do not pass. These exams are known as “extraordinarios”.
+EXAMINATIONS AND MARKS/GRADES	WHEN ARE THE EXTRAORDINARIO EXAMS?	Extraordinario exams for the first term are held in late January or early February. The extraordinario exams for the second term are held in late June.
+EXAMINATIONS AND MARKS/GRADES	I PASSED AN EXAM BUT WITH A LOW MARK/GRADE AND I WANT TO IMPROVE IT. CAN I DO SO?	You should ask your course teacher; some allow you to do an assignment in order to increase your mark/grade; others allow you to take the exam again, others do not allow you to improve your mark. Your teacher must tell you what will happen if they allow you to take the exam again and you fail the second time.You should never take the extraordinario exam without first speaking to your teacher.
+EXAMINATIONS AND MARKS/GRADES	I AM GOING TO TAKE AN EXAM AGAIN. WHAT MUST I DO?	If they are first term exams and you are leaving in the second term, nothing.If they are first term exams and you are leaving in February or if they are second term exams, you should do the following:As soon as you know that you need to take at least one extraordinario exam, send an email to  incoming@uva.es (mailto:incoming@uva.es)  or  chico@uva.es (mailto:raquel.chico@uva.es)  to let us know that you will be doing extraordinario exams.During your final extraordinario exam, you should ask the teacher invigilating for a proof of attendance note (justificante de asistencia).After your final exam, you should take the proof of attendance note to the International Relations Office, or to your coordinating teacher if you are studying in Palencia, Segovia or Soria.
+EXAMINATIONS AND MARKS/GRADES	HOW WILL I RECEIVE MY MARKS/GRADES CERTIFICATE, TRANSCRIPT OF RECORDS, DOCUMENT AFTER THE MOBILITY?	The International Relations Office will send it to you by email.
+EXAMINATIONS AND MARKS/GRADES	MY UNIVERSITY HAS A DOCUMENT FOR THIS MARKS/GRADES CERTIFICATE. SHOULD I TAKE IT TO MY COORDINATING TEACHER TO HAVE IT SIGNED?	No. The rules in our university require us to use our own certificate. This is created following a strict and safe protocol. It also has to be signed by the university authorities, the Vice-rector for International Relations, and the General Secretary. A transcript signed by your coordinating teacher at the University of Valladolid will be not be valid.If this is a problem for your university, you should contact us directly.
+BEFORE LEAVING THE UNIVERSITY OF VALLADOLID	WHAT MUST I DO WHEN I FINISH MY STUDIES AT THE UNIVERSITY OF VALLADOLID?	You should pick up the certificate accrediting your stay (certificado de estancia) or Letter 2 at the same place where you received your certificate of arrival or Letter 1:If you have studied in the city of Valladolid, you should go to the International Relations Office. If you have studied in Palencia, Segovia, or Soria, you should see your coordinating teacher.The International Relations Office will send you your marks/grades certificate or transcript of records directly. 
+EXTENDING AN EXCHANGE STUDY STAY	HOW CAN I EXTEND MY STAY AT THE UVA?	You should fill in the following document ( download here) (http://relint.uva.es/wp-content/uploads/2019/02/DOCUMENTO_AMPLIACION_MATRICULA.pdf)  and send it by email to  incoming@uva.es (mailto:incoming@uva.es)  or hand it in personally at the International Relations Office.
+EXTENDING AN EXCHANGE STUDY STAY	MY UNIVERSITY WILL NOT GIVE ME AN EXTENSION. IS THERE ANOTHER WAY I CAN STAY AT THE UVA?	Yes. Through the Visiting Students Programme, any university student may stay at the University of Valladolid for one or two terms. 

intrel_qa_pairs.json

+[
+    {
+        "answer": "You should find out whether there is an exchange agreement between our universities. To do this, you should contact  incoming@uva.es (mailto:incoming@uva.es) If there is no agreement between our universities, you can always come as a  Visiting student (http://relint.uva.es/internacional/english/students/visitors-programme/) .",
+        "page": "relint",
+        "question": "CAN I DO A STUDY EXCHANGE AT THE UNIVERSITY OF VALLADOLID?",
+        "section": "STUDIES AT THE UNIVERSITY OF VALLADOLID"
+    },
+    {
+        "answer": "You can choose the courses you like from any academic course or academic year. You can also even take courses from various curricula and from different faculties at the same time. However, you must make sure that your coordinating teacher accepts your choices when signing your studies contract or Learning Agreement.Whatever you opt to do, you should take at least one course from the University of Valladolid faculty you have chosen.",
+        "page": "relint",
+        "question": "WHICH COURSES CAN I TAKE AT THE UNIVERSITY OF VALLADOLID?",
+        "section": "STUDIES AT THE UNIVERSITY OF VALLADOLID"
+    },
+    {
+        "answer": "No, you can only choose courses taught during the time you are at our university.If you are here between September and February, you can choose courses taught in the first term or semester.If you are here between February and June, you can choose courses taught in the second term or semester.If you are here from September to June, you can choose any course, including full-year courses.",
+        "page": "relint",
+        "question": "CAN I CHOOSE COURSES FROM ANY TERM OR SEMESTER?",
+        "section": "STUDIES AT THE UNIVERSITY OF VALLADOLID"
+    },
+    {
+        "answer": "No. Full-year courses are taught continuously from September until June. You are not allowed to take them if you are only going to be here during the first or second semester.",
+        "page": "relint",
+        "question": "CAN I CHOOSE FULL-YEAR COURSES IF I STUDY AT THE UVA FOR A TERM OR A SEMESTER?",
+        "section": "STUDIES AT THE UNIVERSITY OF VALLADOLID"
+    },
+    {
+        "answer": "Yes, provided that your coordinating teacher agrees to this when signing your studies contract or Learning Agreement. However, you must always take at least one course from the University of Valladolid faculty you have chosen.",
+        "page": "relint",
+        "question": "CAN I CHOOSE COURSES FROM VARIOUS FACULTIES AT THE SAME TIME?",
+        "section": "STUDIES AT THE UNIVERSITY OF VALLADOLID"
+    },
+    {
+        "answer": "You may only choose courses from the Valladolid and Palencia campuses at the same time, provided that your coordinating teacher agrees to this when signing your studies contract or Learning Agreement. All of the other combinations of campuses are over 100 kilometres away from each other.",
+        "page": "relint",
+        "question": "CAN I TAKE COURSES AT VARIOUS CAMPUSES AT THE SAME TIME?",
+        "section": "STUDIES AT THE UNIVERSITY OF VALLADOLID"
+    },
+    {
+        "answer": "Yes, provided that your coordinating teacher agrees to this when signing your studies contract or Learning Agreement.",
+        "page": "relint",
+        "question": "I AM A MASTER’S DEGREE STUDENT. CAN I TAKE BACHELOR’S DEGREE COURSES?",
+        "section": "STUDIES AT THE UNIVERSITY OF VALLADOLID"
+    },
+    {
+        "answer": "You can see the various bachelor's degree curricula  here (http://www.uva.es/export/sites/uva/2.docencia/2.01.grados/2.01.02.ofertaformativagrados/2.01.02.01.alfabetica/index.html) , and the master's degree curricula  here (http://www.uva.es/export/sites/uva/2.docencia/2.02.mastersoficiales/2.02.01.ofertaeducativa/2.02.01.01.alfabetica/index.html) . Choose the curriculum you are interested in, and then click on \"Asignaturas\" (courses) to see the courses. ",
+        "page": "relint",
+        "question": "WHERE CAN I SEE THE UVA CURRICULA?",
+        "section": "STUDIES AT THE UNIVERSITY OF VALLADOLID"
+    },
+    {
+        "answer": "Each University of Valladolid faculty has its own webpage where the calendars, timetables and classrooms for the courses are published. You can find the link to your centre’s web  here (http://www.uva.es/export/sites/uva/2.docencia/2.01.grados/2.01.02.ofertaformativagrados/2.01.02.01.alfabetica/index.html) .Choose the curriculum you are interested in, and then click on \"Horarios\" (timetables) to see the times and rooms for your courses.",
+        "page": "relint",
+        "question": "WHERE CAN I FIND THE TIMETABLE AND CLASSROOMS FOR MY COURSES?",
+        "section": "STUDIES AT THE UNIVERSITY OF VALLADOLID"
+    },
+    {
+        "answer": "Some courses are divided into groups because not all of the students can fit into the classroom at the same time. Some courses may have more than one group for the lectures, some for practical lessons, and others for both lectures and practical lessons.Exchange students can choose whichever groups suit them best. In other words, they can choose group 1 for the lectures in one course, and group 2 for another, etc. However, once a student has chosen a particular group for a course, they should always go to the same group for that course.",
+        "page": "relint",
+        "question": "IN SOME COURSES IT MENTIONS GROUPS SUCH AS 1 T, 2 A, 1 S. WHAT DOES THIS MEAN? CAN I CHOOSE THE GROUP I WANT?",
+        "section": "STUDIES AT THE UNIVERSITY OF VALLADOLID"
+    },
+    {
+        "answer": "You can take a maximum of 30 ECTS (European Credit Transfer and Accumulation System).We do not set a minimum number of credits. If you need to choose a minimum number of credits you must notify your university.",
+        "page": "relint",
+        "question": "WHAT IS THE MINIMUM NUMBER AND THE MAXIMUM NUMBER OF CREDITS I CAN TAKE?",
+        "section": "STUDIES AT THE UNIVERSITY OF VALLADOLID"
+    },
+    {
+        "answer": "You only need to send it to us if we ask you for it. If we don’t ask you for it, it is because we don’t need it yet. Generally speaking, Erasmus+ students can fill in their Learning Agreement and obtain their coordinating teacher’s signature after they get to the University of Valladolid.",
+        "page": "relint",
+        "question": "WHEN SHOULD I SEND MY LEARNING AGREEMENT OR STUDIES PROGRAMME?",
+        "section": "STUDIES AT THE UNIVERSITY OF VALLADOLID"
+    },
+    {
+        "answer": "You should send it directly to your coordinating teacher at the University of Valladolid.",
+        "page": "relint",
+        "question": "I NEED TO SIGN MY LEARNING AGREEMENT. WHO DO I SEND IT TO?",
+        "section": "STUDIES AT THE UNIVERSITY OF VALLADOLID"
+    },
+    {
+        "answer": "If you have any questions related to academic matters or if you need your Learning Agreement signed, you can contact your faculty’s International Relations Coordinator at the University of Valladolid.For any other kinds of question, contact  incoming@uva.es (mailto:incoming@uva.es) ",
+        "page": "relint",
+        "question": "MY COORDINATING TEACHER ISN’T ANSWERING MY EMAILS. WHO CAN HELP ME?",
+        "section": "STUDIES AT THE UNIVERSITY OF VALLADOLID"
+    },
+    {
+        "answer": "You can find information on the general website  http://www.uva.es/ (http://www.uva.es/) . The enrolment procedure is the same for Spanish students as for students of other nationalities. If you have any doubts, you should contact the  Seccion de Alumnos (http://directorio.uva.es/detalle.html?id=4001004) .",
+        "page": "relint",
+        "question": "I WANT TO DO A FULL BACHELOR’S/MASTER’S/DOCTORAL DEGREE AT THE UNIVERSITY OF VALLADOLID. WHERE CAN I GET INFORMATION?",
+        "section": "STUDIES AT THE UNIVERSITY OF VALLADOLID"
+    },
+    {
+        "answer": "Approximately 90% of the lessons are in Spanish. A few courses are also in English, and there are courses in other languages as foreign languages.",
+        "page": "relint",
+        "question": "WHAT LANGUAGE OF INSTRUCTION IS USED AT THE UNIVERSITY OF VALLADOLID?",
+        "section": "THE LANGUAGE OF INSTRUCTION"
+    },
+    {
+        "answer": "Some degrees have a lot of courses in English, such as the Bachelor's Degree in English Studies. There are also courses taught in English. You can check which ones  here (http://relint.uva.es/wp-content/uploads/2019/03/English18-19.pdf) .You can also take the International Semesters:https://campusvirtual.uva.es/course/index.php?categoryid=309http://www2.emp.uva.es/index.php/relaciones-internacionales-facultad-de-comercio-de-valladolid/https://campusvirtual.uva.es/course/index.php?categoryid=277http://www.feyts.uva.es/?q=node/2424",
+        "page": "relint",
+        "question": "CAN I STUDY IN ENGLISH AT THE UNIVERSITY OF VALLADOLID?",
+        "section": "THE LANGUAGE OF INSTRUCTION"
+    },
+    {
+        "answer": "You need a B1 level of Spanish (Common European Framework of Reference for Languages - CEFR) when you get to the University of Valladolid.",
+        "page": "relint",
+        "question": "WHAT LEVEL OF SPANISH DO I NEED TO STUDY AT THE UVA?",
+        "section": "THE LANGUAGE OF INSTRUCTION"
+    },
+    {
+        "answer": "Yes. We only need to be sure that you won’t have any difficulty understanding your teachers, such that we accept non-official certificates.",
+        "page": "relint",
+        "question": "I HAVE A GOOD COMMAND OF SPANISH BUT DO NOT HAVE AN OFFICIAL B1 CERTIFICATE. CAN I PRESENT OTHER CERTIFICATES OF SPANISH?",
+        "section": "THE LANGUAGE OF INSTRUCTION"
+    },
+    {
+        "answer": "You can send us a report from a Spanish language teacher at your university, or from your International Relations Office, stating that you have sufficient command of Spanish to be able to satisfactorily undertake your stay at our university.Another way to acquire a sufficient level of Spanish is to take an intensive course at a language centre or language school in Spain before you start your academic year.",
+        "page": "relint",
+        "question": "I SPEAK SPANISH BUT DO NOT HAVE CERTIFICATES. WHAT CAN I DO?",
+        "section": "THE LANGUAGE OF INSTRUCTION"
+    },
+    {
+        "answer": "If you are going to study in the city of Valladolid, you should go to the International Relations Office to register that you have arrived and we will give you information. You can find our address and the office opening hours through this link.If you are going to study in Palencia, Segovia, or Soria you should visit your coordinating teacher, who will register your arrival and will give you academic information. You should let your coordinating teacher know in advance by email the date on which you will be arriving. You can find their contact details in your Personal Area.If your coordinating teacher is late in answering you, you can also contact the assistant at the International Relations Office on your campus through the following email addresses:Palencia Campus:  palencia@uva.es (mailto:internacional.palencia@uva.es) Segovia Campus:  segovia@uva.es (mailto:internacional.segovia@uva.es) Soria Campus:  soria@uva.es (mailto:internacional.soria@uva.es) ",
+        "page": "relint",
+        "question": "WHAT DO I HAVE TO DO THE FIRST DAY I GET TO THE UVA?",
+        "section": "WHEN YOU GET TO THE UNIVERSITY OF VALLADOLID"
+    },
+    {
+        "answer": "You can still have access to all university services and facilities even if you are still waiting for your student card.There are two ways to access: with your enrolment sheet or downloading the UVa app for smartphones, available for  Android (https://play.google.com/store/apps/details?id=com.pocketuniversity.uva)  and  iOS (https://itunes.apple.com/app/id1031505989) .",
+        "page": "relint",
+        "question": "I STILL DO NOT HAVE MY STUDENT CARD AND I NEED IT. WHAT SHOULD I DO?",
+        "section": "WHEN YOU GET TO THE UNIVERSITY OF VALLADOLID"
+    },
+    {
+        "answer": "Each country has a different grading system. European countries share a transfer system which you can check out in Wikipedia:  ECTS (https://es.wikipedia.org/wiki/European_Credit_Transfer_and_Accumulation_System) ,  Grading scale (only in English) (https://en.wikipedia.org/wiki/ECTS_grading_scale) .",
+        "page": "relint",
+        "question": "WHAT MARK/GRADE WILL I GET IN MY COUNTRY WITH THE EXAMINATIONS I TAKE AT THE UNIVERSITY OF VALLADOLID?",
+        "section": "EXAMINATIONS AND MARKS/GRADES"
+    },
+    {
+        "answer": "First term examinations are held in January, although some teachers may bring theirs forward to late December.Second term examinations are held in June.",
+        "page": "relint",
+        "question": "WHEN ARE EXAMINATIONS HELD AT THE UNIVERSITY OF VALLADOLID?",
+        "section": "EXAMINATIONS AND MARKS/GRADES"
+    },
+    {
+        "answer": "You should ask your course teacher. They will tell you if there is another way for you to take the exam.",
+        "page": "relint",
+        "question": "I AM NOT IN SPAIN ON THE DAY OF THE EXAM. CAN I TAKE THE EXAM IN MY HOME COUNTRY?",
+        "section": "EXAMINATIONS AND MARKS/GRADES"
+    },
+    {
+        "answer": "In Valladolid, you have the chance to take an exam again if you do not pass. These exams are known as “extraordinarios”.",
+        "page": "relint",
+        "question": "I DID NOT PASS AN EXAM. WHAT CAN I DO?",
+        "section": "EXAMINATIONS AND MARKS/GRADES"
+    },
+    {
+        "answer": "Extraordinario exams for the first term are held in late January or early February. The extraordinario exams for the second term are held in late June.",
+        "page": "relint",
+        "question": "WHEN ARE THE EXTRAORDINARIO EXAMS?",
+        "section": "EXAMINATIONS AND MARKS/GRADES"
+    },
+    {
+        "answer": "You should ask your course teacher; some allow you to do an assignment in order to increase your mark/grade; others allow you to take the exam again, others do not allow you to improve your mark. Your teacher must tell you what will happen if they allow you to take the exam again and you fail the second time.You should never take the extraordinario exam without first speaking to your teacher.",
+        "page": "relint",
+        "question": "I PASSED AN EXAM BUT WITH A LOW MARK/GRADE AND I WANT TO IMPROVE IT. CAN I DO SO?",
+        "section": "EXAMINATIONS AND MARKS/GRADES"
+    },
+    {
+        "answer": "If they are first term exams and you are leaving in the second term, nothing.If they are first term exams and you are leaving in February or if they are second term exams, you should do the following:As soon as you know that you need to take at least one extraordinario exam, send an email to  incoming@uva.es (mailto:incoming@uva.es)  or  chico@uva.es (mailto:raquel.chico@uva.es)  to let us know that you will be doing extraordinario exams.During your final extraordinario exam, you should ask the teacher invigilating for a proof of attendance note (justificante de asistencia).After your final exam, you should take the proof of attendance note to the International Relations Office, or to your coordinating teacher if you are studying in Palencia, Segovia or Soria.",
+        "page": "relint",
+        "question": "I AM GOING TO TAKE AN EXAM AGAIN. WHAT MUST I DO?",
+        "section": "EXAMINATIONS AND MARKS/GRADES"
+    },
+    {
+        "answer": "The International Relations Office will send it to you by email.",
+        "page": "relint",
+        "question": "HOW WILL I RECEIVE MY MARKS/GRADES CERTIFICATE, TRANSCRIPT OF RECORDS, DOCUMENT AFTER THE MOBILITY?",
+        "section": "EXAMINATIONS AND MARKS/GRADES"
+    },
+    {
+        "answer": "No. The rules in our university require us to use our own certificate. This is created following a strict and safe protocol. It also has to be signed by the university authorities, the Vice-rector for International Relations, and the General Secretary. A transcript signed by your coordinating teacher at the University of Valladolid will be not be valid.If this is a problem for your university, you should contact us directly.",
+        "page": "relint",
+        "question": "MY UNIVERSITY HAS A DOCUMENT FOR THIS MARKS/GRADES CERTIFICATE. SHOULD I TAKE IT TO MY COORDINATING TEACHER TO HAVE IT SIGNED?",
+        "section": "EXAMINATIONS AND MARKS/GRADES"
+    },
+    {
+        "answer": "You should pick up the certificate accrediting your stay (certificado de estancia) or Letter 2 at the same place where you received your certificate of arrival or Letter 1:If you have studied in the city of Valladolid, you should go to the International Relations Office. If you have studied in Palencia, Segovia, or Soria, you should see your coordinating teacher.The International Relations Office will send you your marks/grades certificate or transcript of records directly. ",
+        "page": "relint",
+        "question": "WHAT MUST I DO WHEN I FINISH MY STUDIES AT THE UNIVERSITY OF VALLADOLID?",
+        "section": "BEFORE LEAVING THE UNIVERSITY OF VALLADOLID"
+    },
+    {
+        "answer": "You should fill in the following document ( download here) (http://relint.uva.es/wp-content/uploads/2019/02/DOCUMENTO_AMPLIACION_MATRICULA.pdf)  and send it by email to  incoming@uva.es (mailto:incoming@uva.es)  or hand it in personally at the International Relations Office.",
+        "page": "relint",
+        "question": "HOW CAN I EXTEND MY STAY AT THE UVA?",
+        "section": "EXTENDING AN EXCHANGE STUDY STAY"
+    },
+    {
+        "answer": "Yes. Through the Visiting Students Programme, any university student may stay at the University of Valladolid for one or two terms. ",
+        "page": "relint",
+        "question": "MY UNIVERSITY WILL NOT GIVE ME AN EXTENSION. IS THERE ANOTHER WAY I CAN STAY AT THE UVA?",
+        "section": "EXTENDING AN EXCHANGE STUDY STAY"
+    }
+]
\ No newline at end of file

most_similar.ipynb

+{
+ "nbformat": 4,
+ "nbformat_minor": 2,
+ "metadata": {
+  "language_info": {
+   "name": "python",
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "version": "3.6.9-final"
+  },
+  "orig_nbformat": 2,
+  "file_extension": ".py",
+  "mimetype": "text/x-python",
+  "name": "python",
+  "npconvert_exporter": "python",
+  "pygments_lexer": "ipython3",
+  "version": 3,
+  "kernelspec": {
+   "name": "python36964bitvenvvenv9e2469329f0140ba9ee4887b501a6555",
+   "display_name": "Python 3.6.9 64-bit ('.venv': venv)"
+  }
+ },
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 40,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import en_core_web_lg\n",
+    "import numpy as np\n",
+    "import json\n",
+    "\n",
+    "from sklearn.metrics.pairwise import cosine_similarity\n",
+    "\n",
+    "nlp = en_core_web_lg.load()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 41,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "\n",
+    "def get_response(msg):\n",
+    "    \"\"\" Processes a message and generates a respone \"\"\"\n",
+    "    return \"You said '{}'\".format(msg)\n",
+    "\n",
+    "\n",
+    "def answer_question(question):\n",
+    "    \"\"\" Determines the most similar question and provides its answer \"\"\"\n",
+    "    \n",
+    "\n",
+    "def most_similar(question, n=1):\n",
+    "    \"\"\" Obtains the n most similar questins and its similarity to a given\n",
+    "        question \"\"\"\n",
+    "    \n",
+    "    with open('./qa_pairs_em.json', 'r', encoding='utf8') as file:\n",
+    "        qa_pairs = json.load(file)\n",
+    "    \n",
+    "    request_embedding = sentence_embedding(question)\n",
+    "\n",
+    "    for pair in qa_pairs:\n",
+    "        pair['sim'] = cosine_similarity(np.array(pair['embedding']).reshape(1, -1), request_embedding.reshape(1, -1))\n",
+    "\n",
+    "    sorted_qa_pairs = sorted(qa_pairs, key=lambda p: p['sim'], reverse=True)\n",
+    "\n",
+    "    return sorted_qa_pairs[0:n]\n",
+    " \n",
+    "\n",
+    "\n",
+    "\n",
+    "def sentence_embedding(question):\n",
+    "    \"\"\" Computes the embedding os a sentence, removing the punctuation \n",
+    "        symbols and stop words \"\"\"\n",
+    "    doc = nlp(question)\n",
+    "    tokens = list(filter(lambda t: t.has_vector and (not t.is_punct) and (not t.is_stop), doc.doc)) \n",
+    "    vectors = list(map(lambda t: t.vector, tokens))\n",
+    "    embedding = np.average(vectors, axis=0)\n",
+    "    \n",
+    "    return embedding\n",
+    "\n",
+    "    \n",
+    "\n",
+    "    "
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 42,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": "[{'answer': 'You can take a maximum of 30 ECTS (European Credit Transfer and Accumulation System).We do not set a minimum number of credits. If you need to choose a minimum number of credits you must notify your university.',\n  'embedding': [-0.17360201478004456,\n   0.2857059836387634,\n   -0.09179359674453735,\n   -0.03986828029155731,\n   0.09077881276607513,\n   -0.09095919877290726,\n   -0.056734006851911545,\n   -0.08263679593801498,\n   0.14751499891281128,\n   1.6455860137939453,\n   -0.09618699550628662,\n   0.37244200706481934,\n   -0.26668840646743774,\n   -0.01920079067349434,\n   0.3285300135612488,\n   0.1448580026626587,\n   -0.016309011727571487,\n   1.9263579845428467,\n   -0.1175381988286972,\n   -0.14208020269870758,\n   -0.47690001130104065,\n   0.11448699235916138,\n   -0.019088000059127808,\n   -0.07032860070466995,\n   0.020871996879577637,\n   -0.3075757622718811,\n   -0.0240160021930933,\n   -0.19370579719543457,\n   0.09471800178289413,\n   0.19954800605773926,\n   0.24274274706840515,\n   -0.1653279960155487,\n   0.06869599968194962,\n   -0.05831500142812729,\n   -0.047429002821445465,\n   0.21814760565757751,\n   -0.27856120467185974,\n   0.1793222427368164,\n   0.16480259597301483,\n   -0.10944481194019318,\n   0.03280999884009361,\n   0.11687976121902466,\n   0.07233400642871857,\n   -0.2473059892654419,\n   -0.08093379437923431,\n   -0.012797999195754528,\n   0.09415080398321152,\n   0.3226119875907898,\n   0.13129059970378876,\n   0.33777180314064026,\n   -0.09521840512752533,\n   -0.0972302109003067,\n   0.42097002267837524,\n   -0.2178020477294922,\n   -0.35405200719833374,\n   0.2625020146369934,\n   0.18072399497032166,\n   -0.1045634001493454,\n   -0.022241998463869095,\n   0.05760600045323372,\n   0.11243180930614471,\n   -0.018262341618537903,\n   0.041218001395463943,\n   0.28095778822898865,\n   0.18359999358654022,\n   0.1079895943403244,\n   -0.07681719958782196,\n   0.44262999296188354,\n   0.021503198891878128,\n   -0.16929659247398376,\n   0.21874800324440002,\n   -0.25487199425697327,\n   0.24597401916980743,\n   -0.1666940152645111,\n   -0.10225560516119003,\n   0.08862639963626862,\n   0.3300279974937439,\n   0.27038803696632385,\n   0.26868799328804016,\n   0.3069559931755066,\n   0.1084074005484581,\n   0.0025721967685967684,\n   -0.12366900593042374,\n   -0.01648399792611599,\n   0.22017399966716766,\n   0.3083154261112213,\n   -0.12271080166101456,\n   0.0014439880615100265,\n   0.4134979844093323,\n   -0.2176080048084259,\n   -0.1547360122203827,\n   -0.031072799116373062,\n   0.10116299241781235,\n   -0.3913640081882477,\n   -0.2730939984321594,\n   -0.40087199211120605,\n   -0.11903717368841171,\n   0.30919259786605835,\n   -0.08509880304336548,\n   0.2170100212097168,\n   -0.3247320055961609,\n   0.058950792998075485,\n   -0.12724998593330383,\n   0.29962602257728577,\n   0.11356399208307266,\n   -1.3211040496826172,\n   0.11673639714717865,\n   -0.12790599465370178,\n   -0.276995986700058,\n   -0.1455100029706955,\n   0.18558959662914276,\n   -0.031009402126073837,\n   0.21197399497032166,\n   0.29686781764030457,\n   0.2465953826904297,\n   -0.22457000613212585,\n   0.013078195974230766,\n   0.21981699764728546,\n   0.09826799482107162,\n   0.02775081992149353,\n   0.14509379863739014,\n   0.1350167989730835,\n   -0.07278880476951599,\n   -0.18373680114746094,\n   -0.06714359670877457,\n   -0.03427640348672867,\n   -0.0044256048277020454,\n   -0.0026993989013135433,\n   -0.23584198951721191,\n   -0.03793000057339668,\n   0.28535598516464233,\n   0.5475040078163147,\n   0.3007040023803711,\n   0.09356200695037842,\n   0.3532800078392029,\n   0.001216998673044145,\n   0.177620992064476,\n   0.17242200672626495,\n   -0.1471319943666458,\n   0.09812700748443604,\n   -0.29696398973464966,\n   0.23708602786064148,\n   0.12076999992132187,\n   0.08454179763793945,\n   0.11396720260381699,\n   0.011809198185801506,\n   0.12508000433444977,\n   0.09127520024776459,\n   -0.23609140515327454,\n   0.11979091167449951,\n   0.1290608048439026,\n   0.10515180975198746,\n   -0.1819620132446289,\n   -0.029345199465751648,\n   -0.0949850082397461,\n   -0.13489320874214172,\n   0.10540719330310822,\n   0.3035481572151184,\n   -0.08335208892822266,\n   -0.08129139244556427,\n   0.27863600850105286,\n   -0.10348440706729889,\n   0.004668998531997204,\n   -0.1886695921421051,\n   -0.27368199825286865,\n   -0.1477540135383606,\n   0.19394339621067047,\n   -0.20967161655426025,\n   0.05106360465288162,\n   0.03730279952287674,\n   0.13471360504627228,\n   -0.09926401078701019,\n   0.18258564174175262,\n   -0.3133159875869751,\n   -0.24087198078632355,\n   0.0999160036444664,\n   0.04247479885816574,\n   -0.07674960047006607,\n   0.0871339961886406,\n   -0.17373399436473846,\n   0.24380400776863098,\n   0.2539118230342865,\n   0.06291340291500092,\n   0.11732339859008789,\n   -0.19952599704265594,\n   0.07520799338817596,\n   -0.2304907739162445,\n   -0.004106199834495783,\n   -0.10559400171041489,\n   0.27571845054626465,\n   0.27647361159324646,\n   0.239777609705925,\n   -0.006896001286804676,\n   0.1134980097413063,\n   0.1992587149143219,\n   -0.08205579966306686,\n   0.012253999710083008,\n   0.06127599626779556,\n   -0.13951480388641357,\n   0.14521659910678864,\n   0.25714385509490967,\n   -0.5186220407485962,\n   -0.3629215955734253,\n   0.047586195170879364,\n   -0.022932002320885658,\n   0.05594300106167793,\n   0.03171540051698685,\n   0.21590439975261688,\n   -0.06544600427150726,\n   0.13766399025917053,\n   0.2444848120212555,\n   0.34897738695144653,\n   -0.158473402261734,\n   -0.3224756121635437,\n   -0.01880520209670067,\n   0.37882399559020996,\n   -0.22200199961662292,\n   0.03375949710607529,\n   -0.11233800649642944,\n   0.44629397988319397,\n   0.0842847228050232,\n   0.038134194910526276,\n   0.10011999309062958,\n   -0.09800399839878082,\n   0.06253720074892044,\n   -0.03526441007852554,\n   0.09005999565124512,\n   -0.20750999450683594,\n   0.16696040332317352,\n   0.18080200254917145,\n   0.31518638134002686,\n   -0.035266004502773285,\n   -0.0067013963125646114,\n   0.20010599493980408,\n   -0.2820040285587311,\n   0.08720959722995758,\n   -0.027991998940706253,\n   -0.10632185637950897,\n   -0.24111399054527283,\n   0.3901039958000183,\n   -0.09296020120382309,\n   0.07975540310144424,\n   -0.08928759396076202,\n   0.38221877813339233,\n   0.1776427924633026,\n   -0.27977117896080017,\n   0.1826941967010498,\n   0.09810199588537216,\n   -0.08979939669370651,\n   -0.016914403066039085,\n   -0.12272019684314728,\n   0.0751442089676857,\n   -0.04050440713763237,\n   -0.12238439172506332,\n   -0.02614399418234825,\n   0.9553019404411316,\n   -0.17295241355895996,\n   0.01173519529402256,\n   0.2185879945755005,\n   0.09779399633407593,\n   0.18251679837703705,\n   0.35700997710227966,\n   -0.23781640827655792,\n   0.22407200932502747,\n   -0.08763179928064346,\n   -0.3127191960811615,\n   0.3250959813594818,\n   0.1636119931936264,\n   0.7351981997489929,\n   -0.3778315782546997,\n   0.1507820188999176,\n   0.0781099945306778,\n   0.21310201287269592,\n   -0.32892221212387085,\n   -0.019435999915003777,\n   0.03358200192451477,\n   -0.19427600502967834,\n   0.1710543930530548,\n   0.011434400454163551,\n   0.05372900515794754,\n   -0.005374002270400524,\n   0.4516007900238037,\n   -0.020184801891446114,\n   -0.07182079553604126,\n   0.0071439980529248714,\n   -0.04806799441576004,\n   -0.20998802781105042,\n   -0.04649186134338379,\n   0.1629486083984375,\n   0.42056798934936523,\n   -0.15683820843696594,\n   -0.0014477968215942383,\n   0.4137340188026428,\n   0.2308119833469391,\n   0.022228797897696495,\n   0.019019605591893196,\n   -0.15768960118293762,\n   -0.1490384042263031,\n   0.06764759868383408,\n   -0.12349740415811539],\n  'page': 'relint',\n  'question': 'WHAT IS THE MINIMUM NUMBER AND THE MAXIMUM NUMBER OF CREDITS I CAN TAKE?',\n  'section': 'STUDIES AT THE UNIVERSITY OF VALLADOLID',\n  'sim': array([[0.77149119]])}]"
+     },
+     "execution_count": 42,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "most_similar('What is the minimum credists I have to course ?')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ]
+}
\ No newline at end of file