Pers O. Najmalddin on LinkedIn: I showcased my capstone project at University Day 2024 at AUIS. I… | 14 comments (2024)

Got the 4th (last) and most challenging course certificate of the NLP specialization.This is "Natural Language Processing with Attention Models" certificate from DeepLearning.AI.After completing the course I am able to build GPT and other Generative models using encoder-decoder architecture, attention (Transformers) architecture and Text-to-Text transfer model.I have also learnt about BERT and T5 model, and about Hugging face framework.This course is part of Natural language processing specialization program offered through Coursera.

19

4 Comments

Happy to Announce Post - A small accomplishment - completed course 1, Natural Language Processing with Classification and Vector Spaces, of the Natural Language Processing Specialization (Coursera, DeepLearning.AI). Last week I completed last course in this specialization and the quality of that made me force to take the remaining courses in this specialization.I expected the course to have NLP concepts I learnt few years back - like stemming, TD-IF, something on RNN/LSTM etc - but glad the course had nothing of that and totally new content for me. Getting straight into embeddings, vector search and how to optimize it (ex - Locality sensitive hashing). Can't wait to complete remaining two courses of the specialization and then start implementing some of the techniques in working with LLMs.DeepLearning.AI is releasing almost 1 short course every 2 to 3 days and one such course include finetuning a LLM using Lamini. If you are already working on LLMs and RAG architecture - it is worth going through this latest course - Finetuning Large Language Models#llm #genai #Lamini #deeplearningai

19

1 Comment

🚨🚨🚨 New article people! What do you know about LDA??🤔📚 Exploring the Power of Latent Dirichlet Allocation (LDA) in Natural Language Processing 🧐LDA is like a magician that uncovers hidden themes and patterns within textual data. It's a topic modeling algorithm that helps us discover the underlying structure of documents or texts. Whether you're a data scientist, a linguist, or simply curious about how machines can make sense of language, LDA has something intriguing to offer. 🔍 Applications of LDA 🔍LDA has found applications in various fields:📊 Data Science: LDA is commonly used for text mining, information retrieval, and document clustering.📰 Media and Journalism: News organizations use LDA to organize and categorize articles and stories.📈 Finance: LDA can help analyze financial news and reports, providing insights into market sentiment.🔬 Life Sciences: In biology and healthcare, LDA has been used for topic modeling in research articles and medical records.💡 Education: Educational institutions use LDA to analyze student essays and identify common themes and areas for improvement.💼 Business: LDA assists in customer feedback analysis, market research, and competitive intelligence.🌐 Web Content: Content recommendation engines often employ LDA to suggest articles or products based on user preferences.As we continue to generate and consume massive amounts of textual data, the importance of tools like LDA in making sense of this information cannot be underestimated. It's an exciting journey for those passionate about NLP and the ever-evolving world of artificial intelligence.

21

Just completed the course "Natural Language Processing with Classification and Vector Spaces" part of the NLP specialization by DeepLearning.AI where i practiced:a) Performing sentiment analysis on tweets. b) Using vector space models to discover relationships between words, and using PCA to reduce the dimensionality of the vector space.c) Building a translation algorithm using pre-computed word embeddings and locality-sensitive hashing to relate words via approximate k-nearest neighbor search. Looking forward to mastering cutting-edge NLP techniques throughout this specialization.

33

5 Comments

I'm thrilled to share that I've completed the second course in the Natural Language Processing Specialization, "Natural Language Processing with Probabilistic Models," offered by DeepLearning.AI on Coursera.During this course, I embarked on an in-depth journey into the core topics of natural language processing. I explored essential concepts such as edit distance, hidden Markov models, N-gram language models, and word embeddings.One of the highlights of this course was the creation of a spellchecker, which allowed me to effectively correct misspelled words. I leveraged the power of probabilities, minimum edit distance, and dynamic programming to build this valuable tool.The course also introduced me to the world of Markov chains and hidden Markov models. I harnessed these concepts to craft part-of-speech tags for a Wall Street Journal text corpus. By delving into Markov chains, hidden Markov models, and the Viterbi algorithm, I acquired the skills to comprehend and predict word sequences in text.Moreover, I gained valuable insights into N-gram language models, their role in calculating sequence probabilities, and the art of building my own autocomplete language model using a text corpus from Twitter. These skills are indispensable for predicting and suggesting text based on context, greatly enhancing the user experience.The exploration of word embeddings and their profound importance in NLP tasks was another fascinating aspect of the course. I developed a continuous bag-of-words model to create word embeddings from Shakespearean text, harnessing the capabilities of neural networks and data preprocessing. These word embeddings have enriched my understanding of the semantic meaning of words and their versatile applications in various NLP tasks.The knowledge and skills I've gained in this course, encompassing dynamic programming, hidden Markov models, and word embeddings, have empowered me to implement autocorrect, autocomplete, and identify part-of-speech tags for words with efficiency and precision.#naturallanguageprocessing #coursera #nlppractitioner #machinelearning #datascience #deeplearning

14

I am happy to share that I have completed the first course, "Natural Language Processing with Classification and Vector Spaces," from the Natural Language Processing Specialization offered by DeepLearning.AI and Coursera.This course provided me with a comprehensive understanding of the most commonly used machine learning algorithms in NLP tasks, such as logistic regression and the Naive Bayes classifier.Throughout the course, I gained knowledge about the vector space model and its significance in representing words efficiently. I also understood how to optimize memory space utilization by extracting rich features and minimizing model training time while maximizing model quality. Additionally, I learned how to utilize pre-trained word embeddings and vector spaces to establish word analogies using cosine similarity. I also discovered how to map higher-dimensional embeddings into lower-dimensional spaces to visualize word clusters.Moreover, the course covered the construction of a simple machine translation system by minimizing the squared Frobenius norm. I also learned about the application of locality-sensitive hashing in approximate k-nearest neighbors (KNN) to accelerate the search process for finding similar translations.Overall, this course has provided me with a strong foundation in natural language processing, enabling me to apply various techniques and algorithms to solve NLP problems effectively.#NLP #MachineLearning #DataScience

6

1 Comment

Successfully completed a project on Sequential Natural Language Processing 🏆, May'23.#rnn#wordembedding#lstm #glove#classification #tensorflow #project #universityoftexas #greatlearning #mygreatlearning #greatlakes

21

1 Comment

I am thrilled to share that I have recently completed the Natural Language Processing with Classification and Vector Spacescertification from coursera!what I've learned:Use logistic regression, naïve Bayes, and word vectors to implement sentiment analysis, complete analogies & translate words.

6

🚀 Exciting News! 🚀I am thrilled to share that I have successfully completed the Natural Language Processing with Classification and Vector Spaces on Coursera! 🎓This course has been an incredible journey, providing deep insights into Natural language Processing, Sentiment Analysis, Logistic Regression, Naive Bayes Classification, Vector Space Model, Machine Translation, Local Sensitive Hashing, Approximate KNN. I want to express my gratitude to Younes Bensouda Mourri, Instructor of AI at Stanford University and Łukasz Kaiser, Staff Research Scientist at Google Brain and Chargé de Recherche at CNRS for their exceptional guidance for offering such a valuable learning experience.📚 Key Takeaways:Sentiment Analysis using Logistic Regression and Naive BayesVector Space Model and Machine TranslationLocal Sensitive Hashing and Approximate KNNHyperplane and Vector Embedding🌐 About the Course:Natural Language Processing with Classification and Vector Spaces is one of the course of Natural Language Processing Specialization. The lab assignments are very good, doesn't use inbuild function for different machine learning algorithm, hence are very good for fundamental concept.🏆 Accomplishment Unlocked:With this achievement, I am more equipped and inspired to leverage NLP in my professional journey. The knowledge gained is not just a skill set but a catalyst for innovation. It opens doors to contribute meaningfully in fields like Machine Translation, Sentiment Analysis, Vector space model, propelling my career to new heights.🔗 Certificate Link:https://lnkd.in/gJUmkEAmI am excited to apply my new found knowledge and skills in my profession. Here's to continuous learning and growth! 🌱#Coursera #LearningJourney #ContinuousEducation #ProfessionalDevelopment #AchievementUnlocked

18

1 Comment

Excited to share that I have successfully completed the 'Natural Language Processing with Classificationand Vector Spaces' course offered by DeepLearning.AI! This course has equipped me with the skills to implementsentiment analysis,complete analogies, and translate words using techniques such as logistic regression, naive Bayes, and word vectors. I have gained valuable insights onmachine translation,word embeddings, locality-sensitive hashing, sentiment analysis, andvector space models.

40

8 Comments