Mohammed Bahageel

مكة جدة السعودية معلومات الاتصال
٤ آلاف متابع أكثر من 500 زميل

انضم لعرض الملف الشخصي

نبذة عني

Experienced data science professional with over 5 years of expertise in artificial…

مقالات Mohammed

عرض كل المقالات

الإسهامات

انضم الآن لعرض كل الإسهامات

الخبرة والتعليم

  • Dr.Samir Abbas Hospital - مستشفى الدكتور سمير عباس

عرض خبرة Mohammed الكاملة

تعرّف على المسمى الوظيفي للأشخاص ومعدل بقائهم في العمل والكثير غير ذلك.

أو

بالنقر على الاستمرار للانضمام أو تسجيل الدخول، فأنت توافق على اتفاقية المستخدم واتفاقية الخصوصية وسياسة ملفات تعريف الارتباط على LinkedIn.

التراخيص والشهادات

المشروعات

  • Doctor AI Assistant based on RAG part one of Generative AI projects

    -

    The Doctor AI Assistant is an advanced artificial intelligence tool designed to aid physicians in diagnosing diseases swiftly and accurately. It provides comprehensive support by addressing a wide array of queries, including those related to ICD10 codes, diagnoses, symptoms, and differential diagnoses across all medical specialties. Additionally, it assists in the submission of relevant insurance claims and ensures adherence to drug indications consistent with ICD10 codes, guidelines, and best…

    The Doctor AI Assistant is an advanced artificial intelligence tool designed to aid physicians in diagnosing diseases swiftly and accurately. It provides comprehensive support by addressing a wide array of queries, including those related to ICD10 codes, diagnoses, symptoms, and differential diagnoses across all medical specialties. Additionally, it assists in the submission of relevant insurance claims and ensures adherence to drug indications consistent with ICD10 codes, guidelines, and best medical practices. With multilingual capabilities, it offers assistance in all languages spoken worldwide, empowering healthcare professionals with unparalleled efficiency and accuracy in patient care. I developed this app as a part of my experiments using Retrieval Augmented Generation.
    step1: I uploaded 25 Medical Handbooks from prestigious medical universities like oxford in a wide range of fields like differential diagnosis , neurology , cardiology , pediatrics etc.
    step2 : I used google Colab to split the uploaded pdf files into chunks with 1000 words chunks size to facilitate the tokenization and chunk overlap of 200 to enhance the context
    step 3: I used OpenAI embeddings to convert the text into vector embeddings with 1536 dimensions
    step4: I created Qdrant client, instantiated the cluster and created the collection to store the text embeddings in points within the collection
    step5: using langChain frame I created a retrieval chain and history aware retriever to develop a context aware application that is capable of remembering the whole context of the conversation
    finally I used Streamlit to build the front end of the application and provide chatbot like GUI

  • Retrieval Augmented Generation (RAG)Part one Chat with Your PDF Files

    -

    Retrieval-augmented generation (RAG) is a technique for enhancing the accuracy and reliability of generative AI models with facts fetched from external sources.
    In other words, it fills a gap in how LLMs work. Under the hood, LLMs are neural networks, typically measured by how many parameters they contain. An LLM’s parameters essentially represent the general patterns of how humans use words to form sentences.
    In this project I explored the possibility of applying RAG to develop…

    Retrieval-augmented generation (RAG) is a technique for enhancing the accuracy and reliability of generative AI models with facts fetched from external sources.
    In other words, it fills a gap in how LLMs work. Under the hood, LLMs are neural networks, typically measured by how many parameters they contain. An LLM’s parameters essentially represent the general patterns of how humans use words to form sentences.
    In this project I explored the possibility of applying RAG to develop conversational AI apps that retrieve knowledge from knowledge base thus augments the knowledge of LLMs and prevent them from hallucinating , or giving factually inaccurate answers , RAG coupled with vector databases has the potential to revolutionize the process of developing AI applications as I mention in my Article Unlocking the Power of Vector Embedding Databases: A Game-Changer in AI Applications in this link https://www.linkedin.com/pulse/unlocking-power-vector-embedding-databases-ai-mohammed-bahageel-nidhf/ please have a look at my project and feel free to contact me

  • Data Visualization as Art with Plotly and Dash libraries

    -

    In this Project I wanted to accomplish an added interactivity by using locations on choropleth map as Input to the call back function so that whenever a user clicks on a map the rest of the charts get updated accordingly, this is an extremely important feature in data visualization, we see in my Microsoft power bi, Business intelligence analysts need this feature the most , points on the map such as locations codes are used as an input to to the call back function in this way :

    in this…

    In this Project I wanted to accomplish an added interactivity by using locations on choropleth map as Input to the call back function so that whenever a user clicks on a map the rest of the charts get updated accordingly, this is an extremely important feature in data visualization, we see in my Microsoft power bi, Business intelligence analysts need this feature the most , points on the map such as locations codes are used as an input to to the call back function in this way :

    in this case we use map as input and the component property here ‘clickData’ and the output is the chosen data in a form of a children that will go into html.Div, the function uses these two to filter the data based on the clicked location

    عرض المشروع

  • Covid 19 Dashboard with Real Time Data Using Plotly and Dash

    -

    concept: This dashboard extracts data from a data source using URL , thus it gets automatically updated as the data is updated in the data source enabling the user to gain real-time insights into the trajectory of the pandemic , off course one might argue that covid 19 pandemic is over , but I would to reiterate the fact that is just a demonstration as to how to apply plotly and dash to build interactive dashboards for all purposes , dash library is built on top java script , flask and react JS…

    concept: This dashboard extracts data from a data source using URL , thus it gets automatically updated as the data is updated in the data source enabling the user to gain real-time insights into the trajectory of the pandemic , off course one might argue that covid 19 pandemic is over , but I would to reiterate the fact that is just a demonstration as to how to apply plotly and dash to build interactive dashboards for all purposes , dash library is built on top java script , flask and react JS , as well as it has HTML and BOOTSTRAP components which makes it an extraordinarily powerful tool for designing dashboards which fit all setting including mobile phone screens and thus making it user friendly , basically I used plotly graph objects to build the charts and then create the layout i.e. the rows and columns in the web page , place those charts within these columns and add interactivity to my dashboard I have defined a set of call back functions that link the charts with interactive components input in my case the dropdown menu , in a nutshell it just filters the data that feeds the chart figures

    Conclusion: I would like to conclude by stating the fact that data science is an enormous field with big potential and data visualization is only one aspect of it , using plotly graph objects along with dash enables a data scientist to draw a picture that worths a thousand words

    عرض المشروع

  • TIME SERIES FORECASTING WITH PYTHON FORECASTING STOCK PRICE

    -

    Concept: This application uses stock market data by applying ticker symbols , the user has the liberty to enter whatever ticker and he or she will get up to date data related to the stock as well as News and forecast the future price , the application relays on a Machine Learning algorithm called Facebook Prophet developed originally by META as well as streamlit library as a framework Within which the application is designed , streamlit is incredibly powerful and advantageous library that…

    Concept: This application uses stock market data by applying ticker symbols , the user has the liberty to enter whatever ticker and he or she will get up to date data related to the stock as well as News and forecast the future price , the application relays on a Machine Learning algorithm called Facebook Prophet developed originally by META as well as streamlit library as a framework Within which the application is designed , streamlit is incredibly powerful and advantageous library that offers the possibility for data scientists and app developers to develop and deploy their Data science , Machine Learning and Computer Vision Application with ridiculous ease , the data in this app is scraped from yahoo finance website using yahoo finance library as well as pandas_ datareader To extract financial data instantaneously which can be used to analyze and capture historical trends and tap into models like Facebook Prophet , Prophet is used in many applications across Facebook for producing reliable forecasts for planning and goal setting. I've found it to perform better than any other approach in the majority of cases Prophet is a procedure for forecasting time series data based on an additive model where non-linear trends are fit with yearly, weekly, and daily seasonality, plus holiday effects. It works best with time series that have strong seasonal effects and several seasons of historical data. Prophet is robust to missing data and shifts in the trend, and typically handles outliers well. It works very much like Sklearn models where you train and deploy the model

    Conclusion: I would like to conclude by emphasizing the fact the Facebook Prophet as model cannot be used in all situations particularly when it comes to forecasting stock prices as there are Many intervening factors beyond the seasonality that are incredibly hard for the model to capture such as sudden changes in monetary policy by a central bank this might spook the markets

    عرض المشروع

اللغات

  • English

    -

  • German

    -

  • French

    -

  • Arabic

    -

  • Italian

    -

التوصيات المستلمة

شخصان قدما توصية لـMohammed

انضم الآن لعرض

عرض ملف Mohammed الشخصي الكامل

  • مشاهدة الأشخاص المشتركين الذين تعرفهم
  • تقديم تعارف
  • تواصل مع Mohammed مباشرة
انضم لعرض الملف الشخصي الكامل

ملفات شخصية أخرى مشابهة

استكشاف مزيد من المنشورات

أعضاء آخرون يحملون اسم ⁦⁩Mohammed Bahageel

4 أخرى باسمMohammed Bahageel على LinkedIn

استعرض الأعضاء الآخرين الذين يحملون اسم ⁦⁩Mohammed Bahageel

اكتسب مهارات جديدة من خلال هذه المواد الدراسية

عرض كل الدورات