--- title: Chatbot emoji: 🌖 colorFrom: gray colorTo: purple sdk: gradio sdk_version: 4.18.0 app_file: app.py pinned: false --- # Repository with chatbot on five algorithms: TF-IDF, W2V (Glove-25), BERT, Bi_BERT, Bi-Cross-BERT ## Introduction This repo was created to compare different retrieval algorithms on chatbot which speaks my favorite character. The favourite character is taken as Rachel Green from Friends. All transcripts were taken from this website "https://fangj.github.io/friends/" For this chatbot *parser.py* file was created to download data from website, parse it with beautifulsoup and create dataframes with certaing constraint, like name of favorite character and threshold of frequency most popular characters. In total about 6k lines of dialogs were collected. For data with labels this amount increased by two times - 12k lines. First part of training based in this notebook [HW1_NLP_GEN_TFIDF_W2V_BERT_StarodubovKG.ipynb](https://huggingface.co/spaces/StKirill/chatbot/blob/main/HW1_NLP_GEN_TFIDF_W2V_BERT_StarodubovKG.ipynb) TF-IDF, W2V, BERT algorithms are considered inside. For cleaning and preparing data for first two algorithms nltk library is used. From the data punctuation and stopwords are removed, also lemmatization is used to short form. Chart below shows that length of sentences in data not exceeds 60 words. !["Length of sentences in data"](data:image/png;base64,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) Chatbot with TF-IDF convert "question" to vector, find equivalent or most relevant vector in database and for this vector extract answer which sends to user. Experiments shows that TF-IDF gives good results from the box. ChatBot with TF-IDF works perfect if we have all possible questions and answers. Second algorithm is W2V. Experiments shows that better to use pretrained vectors instead of trained on small amount of data. This is why "glove-twitter-25" is used. Chatbot with W2V shows, probably, similar results as TF-IDF, but spends less time on processing. It is related to length of vectors, 25 in W2V against 5024 in TF-IDF. 5024 is the length of dictionary in TF-IDF. Third algorithm which could be used with chatbot is BERT. BERT-like models are good because in addition to returning a contextualised embedding for each token. So for sentence all embeddings shoud be summed As BERT model are trained for CLS token, chatbot could find more relevant question from database to user's request. Which leads to better answer, but not always. This chatbot uses "distilbert/distilbert-base-uncased" model from huggingface. Distiled version of BERT increase refence speed by reason of less weights in model. All three methods above don't use training in common understanding, but all vectorize question and find relevant from database. These methods show good results, but don't shows flexibility in answers. They usually return answer which corresponds to database. Last two algorithms are shown in "HW1_NLP_GEN_BI_Cross_StarodubovKG.ipynb". Fourth method is Bi-Encoder. This algorithm requires preliminary training to make model more close to training data. During this training chatbot starts "to speak" more like Rachel. Training of the model shows below: !["Training Bi Encoder"](data:image/png;base64,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) After training model could find the range of possible answerss and return the one with best score *argmax()* To choose better one more precisely from the previous range, chat bot uses Cross-Encoder model, which helps select answer, which fits more to the question. Training of the model shows below !["Training Bi Encoder"](data:image/png;base64,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) To sum up, for the baseline first three algorithms could be used. Chatbot answer with quite dirict answers, which are better than better the dataframe. With last two algorithms, Bi and Cross Encoder, model doesn't use database as dictionary. It tries to find more relevant question to answer with Bi-encoder and filter the best one with Cross-Encoder. ## Architecture * models - contains weights for Bi and Cross encoder models. * parse.py - parser of website ("https://fangj.github.io/friends/"), which download transcripts of friends movie and make csv file * HW1_NLP_GEN_TFIDF_W2V_BERT_StarodubovKG.ipynb - training and tests of TF-IDF, W2V, BERT algorithms * HW1_NLP_GEN_BI_Cross_StarodubovKG.ipynb - training and tests of Bi and Cross Encoder * rachel_friends.csv - dataframe for TF-IDF, W2V, BERT models, which used as dictionary for chatbot * rachel_friends_label.csv - data for Bi and Cross encoder. With labels models tries to identify how speaks rachel. * bi_bert_answers.npy - numpy array with vectorized answers, which are used to faster inference of the model. ## Example of conversation with chatbot Example 1 ![Example 1](example_1.png) Example 2 ![Example 2](example_2.png) Example 3 ![Example 3](example_3.png) Example 4 ![Example 4](example_4.png) Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference