Filter
Exclude
Time range
-
Near
Did you know the right text analytics tools can save you hours of work? Our Text Analytics collection on Microsoft Marketplace has you covered with APIs for sentiment analysis, keyword extraction, and more! #TextAnalytics #APIs
1
1
Transform messy unstructured data into structured, actionable intelligence with AI-powered text analysis solutions. 3rdisearch.com/retina #AI #TextAnalytics #NLP #EnterpriseSearch
2
Unlock insights hidden inside unstructured text data with advanced AI-powered analysis. Transform how you understand enterprise data. 3rdisearch.com/retina #AI #TextAnalytics #NLP #EnterpriseSearch
7
📚🤖 Final results from my Sherlock Holmes NLP Classification Project! Using Python, NLP, and Machine Learning, I trained a model to distinguish The Hound of the Baskervilles from other Sherlock Holmes texts. #NLP #MachineLearning #DataScience #AI #TextAnalytics #SherlockHolmes
1
11
📚🔍 What can Data Science reveal about Sherlock Holmes? Using Python, NLP, and text analytics, I analyzed the complete Sherlock Holmes text collection and built a dashboard with 12 literary insights. #Python #NLP #DataScience #TextAnalytics #SherlockHolmes #MachineLearning #AI
1
9
Alex Smirnoff shares insights on text analytics! Sometimes, traditional ML is the most cost-effective solution. Language models can further enhance these methods. Listen to the full show: youtu.be/4B5fNOtNj4c?si=9jW6… #TextAnalytics #MachineLearning #AI
2
31
AI-generated summaries turn thousands of responses into executive-ready insights. #AIInsights #TextAnalytics #Leadership (1/1)
2
2
29
Analicé con Grok las interacciones de esta señora y esto encontré: - La tendencia general de su conversación es consistente con los posts proporcionados: agresiva, crítica y politizada. - Crítica al gobierno actual (Morena, Claudia Sheinbaum, AMLO y familia). Alrededor del 70% de los posts son ataques políticos. - Comparó mi interacción con otros posts y encontró que el tono negativo persiste: negatividad en ~80% de los posts, con sarcasmo y descalificaciones como norma. - En general los posts no presentan argumentos estructurados; en su lugar, dependen de falacias lógicas para contraatacar. - Es posible que sea insegura, aunque no hay argumentos en el análisis de textos concluyentes, pero por mi experiencia, es muy probable por la baja positividad y el exceso de insultos en sus interacciones (carencia de argumentos, exceso de falacias). - Como Grok es un gran modelo de lenguaje, solo se aplicaron heurísticas para el análisis (conteo básico de palabras y memoria del modelo) ¿Qué opinan? #AnálisisX #PersonalidadEnRedes #TextAnalytics #NaturalLanguageProcessing
Es un Psicólogo frustrado, hay que entenderlo... frustrado y con mucha imaginación.
4
4
9
644
Did you know that emojis can enrich text analytics? 📊 Check out this post from SAS' @cjdinger for tips on how you can work with emojis in your SAS programming 2.sas.com/60114sK63 #AI #TextAnalytics #WorldEmojiDay
4
6
837
𝗗𝗮𝘆 𝟮𝟯 𝗼𝗳 𝗦𝘂𝗺𝗺𝗲𝗿 𝗜𝗻𝘁𝗲𝗿𝗻𝘀𝗵𝗶𝗽 𝗣𝗿𝗼𝗴𝗿𝗮𝗺 𝗼𝗻 𝗠𝗮𝗰𝗵𝗶𝗻𝗲 𝗟𝗲𝗮𝗿𝗻𝗶𝗻𝗴 & 𝗗𝗲𝗲𝗽 𝗟𝗲𝗮𝗿𝗻𝗶𝗻𝗴 #Stanley #SummerInternship2025 #MachineLearning #DeepLearning #NLP #Tokenization #WordEmbeddings #TextPreprocessing #Word2Vec #GloVe #TextAnalytics
2
5
45
🛑 Text Preprocessing Fails: 3 Mistakes That Ruin Your NLP Models You’ve trained a fancy BERT model… but it’s performing like a magic 8-ball. Here’s what probably went wrong before you even fed it data: 1. The Stopwords Trap Mistake: Blindly removing all stopwords. Why It Bombs: "Not bad" → "bad" (negation lost). Fix: Use domain-specific stopwords (e.g., keep "no", "never" for sentiment). 2. Lemmatizing Like a Robot Mistake: Lemmatizing "meeting" → "meet" in legal docs ("Board meeting" ≠ "to meet"). Fix: POS-tag first (`spaCy`’s `token.pos_` saves lives). 3. The Unicode Horror Show Mistake: Assuming UTF-8 handles everything. Nightmare Fuel: "Café" → "Caf√©" → model confusion. Fix: `ftfy` library normalize emojis (e.g., 👍 → ":thumbs_up:"). 💡 Pro Tip: Always inspect 100 samples post-cleaning. If you can’t understand the text, neither can your model. What’s your text prep war story? (I once spent 3 hours debugging… turns out the CSV was Excel-mangled.) #NLP #DataScience #TextAnalytics #MachineLearning #Python
2
80