Tweets Analysis - Keyword: @nathanbenaich
Overview
Total number of tweets analysed
170
Earliest tweet was on
2023-02-08
Latest tweet was on
2023-02-18
Tweets covering
10 days
Average age of authors' accounts
9 years
Summarization
Nathan Benaich discusses the challenge of how to share great machine learning (ML) innovations in medicine due to the lack of public datasets and no way to share internal data. He then looks at ways to apply ML specifically in AI in surgery, with various authors having published works relevant to this space. Benaich also addresses Generative AI and how it will change the way we do biology, and the applications in diagnostic imaging and surgical-data science. Furthermore, he acknowledges the early use of CV for life sciences problems and how it could be used to solve other problems. Finally, he suggests one option to address the lack of publishing opportunities in industry with applied research people being unable to publish due to contract terms is to pick a topic outside their daily job scope.
Topic Modeling
- Machine learning and abstraction
- AI in Medicine and applications
- Generative AI and its applications
- Computer vision problems
- ML solutions and challenges
Emotional Analysis
The tweets in this sample express a variety of emotions, from enthusiasm and optimism about the potential of machine learning and generative AI to frustration and concern about the lack of public datasets and contractual restrictions on publishing. There is also a sense of curiosity and discovery as the conversation explores the possibilities of computer vision and other AI applications. Overall, the tweets reflect an eagerness to explore and learn more about the potential of AI and ML technologies.
Trend Analysis
- Machine Learning (ML)
- Medical Imaging
- Generative Artificial Intelligence (AI)
- Pattern Recognition Problems
- Computer Vision (CV)
Types of Tweets
Number of Retweets
38
Percentage of total tweets
22%
Number of Original tweets
11
Percentage of total tweets
6%
Number of tweets that contain Mentions
170
Percentage of total tweets
100%
Number of tweets that were Replies
121
Percentage of total tweets
71%
Number of tweets that were Quotes
9
Percentage of total tweets
5%
Number of tweets that contain Hashtags
14
Percentage of total tweets
8%
Top 5 devices used to tweet
| Source | Count |
|---|---|
| Twitter Web App | 72 |
| Twitter for iPhone | 60 |
| Twitter for Android | 23 |
| mem.ai | 5 |
| Buffer | 4 |
What devices were used to tweet
Top 10 accounts with highest followers count
| Username | Name | Bio | Followers count |
|---|---|---|---|
| SpirosMargaris | Spiros Margaris | #VC | No. 1 #Fintech @Refinitiv & @Onalytica | @TEDx | @natechsa @ai_mediastalker @GenTwoAG @investwithCARL @SparkLabsGlobal @HeradoHQ | 127,685 |
| InvestmentTalkk | Conor Mac | Here to learn, connect with investors & discuss the stock market. Writing at @TheITNewsletter | 50,157 |
| memdotai | Mem | Your self-organizing workspace.Get Mem It for Twitter: https://t.co/PXu0Fqq8k1…Website: https://t.co/XSDVK00mbv Community: https://t.co/arZa1ml8IS | 39,770 |
| darian314 | Darian Shirazi | GP @GradientVC; tweets may be AI generated | 24,182 |
| VeritasArdentur | Fortis Est Veritas | Absolutely vicious if prodded into an aggressive situation. Sedikit sengklek, couldnt care less. Disclaimer: Ava akibat FaceApp belaka. | 23,338 |
| hkanji | Hussein Kanji | Build more tech. Eat more cake. | 22,748 |
| vidtranslator | VidTranslator - targum.video AI bot | HOWTO Reply with `@vidtranslator translate` to any tweet with a video. Please consider a donation to keep the bot alive https://t.co/RbDkBxgfG1 | 16,136 |
| johnthornhillft | John Thornhill | Innovation editor and tech columnist @FinancialTimes. Founder of @Siftedeu. Ex-deputy editor, Moscow correspondent etc | 10,964 |
| tedcooke | Ed Cooke | Creating fun things: Language learning: https://t.co/HYtfFBjljC Spherical concert-halls: https://t.co/hZtRrDFsgR Online parties: https://t.co/V1Z5g2DbiL | 10,554 |
| dksf | DK | startup storyteller, nostr/bitcoin student, early googler, always curious, nostr:#npub1kuy0wwf0tzzqvgfv8zpw0vaupkds3430jhapwrgfjyn7ecnhpe0qj9kdj8 | 9,420 |
Top 10 accounts with highest friends count
| Username | Name | Bio | Followers count |
|---|---|---|---|
| SpirosMargaris | Spiros Margaris | #VC | No. 1 #Fintech @Refinitiv & @Onalytica | @TEDx | @natechsa @ai_mediastalker @GenTwoAG @investwithCARL @SparkLabsGlobal @HeradoHQ | 20,925 |
| communicable | Elena Yunusov | Head of Marketing @BorealisAI / RBC, Founder, #CryptoCamp⚡️Tech, startups, blockchain, AI | Here to build a future that I’m from | 6,864 |
| hkanji | Hussein Kanji | Build more tech. Eat more cake. | 6,779 |
| shigekzishihara | Shigekazu Ishihara | I am researching/teaching/consulting Kansei Engineering, Ergonomics, Gerontechnology, Applied Psychology, AI & Biomechanics at Hiroshima International Univ | 6,616 |
| ayirpelle | priya joseph | geek, entrepreneur, 'I strictly color outside the lines!', opinions r my own indeed. @ayirpelle , universal handle at this time | 5,003 |
| PaulExcell | Paul Excell | Award Winning Coach, Chair, Board Advisor, Apprentice #StayCurious #SocialMobility #NetPositive #Innovation #STEAM #ScaleUp #7Cs #Leadership CEng FIET FBCS ACC | 5,000 |
| johnthornhillft | John Thornhill | Innovation editor and tech columnist @FinancialTimes. Founder of @Siftedeu. Ex-deputy editor, Moscow correspondent etc | 4,810 |
| vijaybolina | Vijay Bolina | Thinking about safety, security & privacy of AGI at @Google @DeepMind. ex @Mandiant @BoozAllen. tweets my own. | 4,711 |
| commandodev | Ben Ford | Organisational Engineering as a service for scaling tech companies | OODA | Former Royal Marine Commando | ADAPT. RESPOND. WIN! | 3,839 |
| PMOHANSRINIVAS1 | P.MOHANSRINIVAS | SENIOR DATA ENGINEER | 3,392 |
Most active users
| Username | Bio | Number of tweets |
|---|---|---|
| memdotai | Your self-organizing workspace.Get Mem It for Twitter: https://t.co/PXu0Fqq8k1…Website: https://t.co/XSDVK00mbv Community: https://t.co/arZa1ml8IS | 5 |
| ayirpelle | geek, entrepreneur, 'I strictly color outside the lines!', opinions r my own indeed. @ayirpelle , universal handle at this time | 4 |
| yacineMTB | Surfing the stacked fractal super sigmoid. see you on /Ω/ & /λ/. swe @ stripe prev @ auth0 ‘terminally plugged in’ | 3 |
| InvestmentTalkk | Here to learn, connect with investors & discuss the stock market. Writing at @TheITNewsletter | 3 |
| future_of_AI | #AI Essentials publishes only the most insightful links shared by trusted #artificialintelligence, #machinelearning experts helping you focus on what matters. | 3 |
| TechGodArisen | The Angel Of All Things Science & Technology In This World, Awoken. | 3 |
| trunghlt | CEO, Founder of @ActableAI. On the mission to help the world get easy access to the latest AI for Data Analytics & Optimization. | 2 |
| thomasrorystone | Investing in advanced technologies for people and planet @kintsugiad 🏞️ Previously Co-founder of PredictionIO 🐸 (Acquired by Salesforce) and Ph.D. @UCLCS 🖥 | 2 |
| luisortiz345 | Bored, logically speaking, and hard too convince | 2 |
| clemnt | Investing in 🐐 founders https://t.co/n8SDgBw0TS (focus on pre-seed/seed B2B software) Side #SaaS project: @Doctripath | 2 |
Tweets per day
Top 10 tweets with highest Retweet count
| ID | Text | Retweet count |
|---|---|---|
| 1624505707713425408 | @nathanbenaich In ML we often take an application-specific problem and elevate it to a general technical problem. Solving the general problem can still be in service to the specific applied one, and helps connect other general technical solutions to the specific one. Abstraction is good! | 5 |
| 1624582838594031617 | @nathanbenaich Similarities in complexity & complex adaptive systems. See also scaccadic control of vision (Jane Wang at Cornell). Also, not all entities are vertical. See SK, Korea. There is a lot of cross-pollination in many areas and sometimes it takes pressing-in from different sides to… https://t.co/dhQ6ebADnQ | 4 |
| 1625030012343549953 | Yes, #GenAI is going to change everything but not in the way you think https://t.co/2725PIKUwk #fintech #AGI #GenerativeAI #ArtificialIntelligence #AI #MachineLearning #DeepLearning @nathanbenaich @Siftedeu https://t.co/sUhGmEuBWi | 3 |
| 1624470067198189568 | @nathanbenaich This is a common problem for ML in medicine. Great ML innovations happen, but there is often no great public datasets and there is no way for them to share their internal data. So you have to either choose to 1. Not publish (standard route) 2. Publish on non bio data | 3 |
| 1624477630580899840 | @nathanbenaich These methods specifically seem most applicable to AI in surgery, i.e. alignment of surgical videos The last author has recently published work in this space, see https://t.co/n8311wgI5X | 2 |
| 1623328905628098560 | Great talk from @nathanbenaich at the @ai_ucl on building #AI first companies. @UCL_DARK #GenerativeAI #LLMs https://t.co/xjT3dNVXNB | 2 |
| 1624471232384040960 | @nathanbenaich Because so much of bio involves CV? Pathology, X-ray, CT, MRI, etc… In fact medical imagine is probably the biggest use case or profitable CV rn. | 1 |
| 1625392146470535169 | Yes, GenAI is going to change everything — but not in the way you think https://t.co/poCdWUKOH9 #AI #Business via @nathanbenaich | 1 |
| 1624473034005311489 | @GabrielAsher02 @nathanbenaich Not to mention that almost any problem can be framed as a computer vision task. | 0 |
| 1624511490567901185 | @nathanbenaich Only explanation I can think of is to solve the replication of experiments problem in bio? | 0 |
Top 10 tweets with highest Like count
| ID | Text | Like count |
|---|---|---|
| 1624505707713425408 | @nathanbenaich In ML we often take an application-specific problem and elevate it to a general technical problem. Solving the general problem can still be in service to the specific applied one, and helps connect other general technical solutions to the specific one. Abstraction is good! | 133 |
| 1624470067198189568 | @nathanbenaich This is a common problem for ML in medicine. Great ML innovations happen, but there is often no great public datasets and there is no way for them to share their internal data. So you have to either choose to 1. Not publish (standard route) 2. Publish on non bio data | 54 |
| 1624477630580899840 | @nathanbenaich These methods specifically seem most applicable to AI in surgery, i.e. alignment of surgical videos The last author has recently published work in this space, see https://t.co/n8311wgI5X | 15 |
| 1625030012343549953 | Yes, #GenAI is going to change everything but not in the way you think https://t.co/2725PIKUwk #fintech #AGI #GenerativeAI #ArtificialIntelligence #AI #MachineLearning #DeepLearning @nathanbenaich @Siftedeu https://t.co/sUhGmEuBWi | 13 |
| 1624471232384040960 | @nathanbenaich Because so much of bio involves CV? Pathology, X-ray, CT, MRI, etc… In fact medical imagine is probably the biggest use case or profitable CV rn. | 11 |
| 1623328905628098560 | Great talk from @nathanbenaich at the @ai_ucl on building #AI first companies. @UCL_DARK #GenerativeAI #LLMs https://t.co/xjT3dNVXNB | 10 |
| 1624596003344076801 | @nathanbenaich woah the same models we use to translate chinese to english work on vision? and they work on robot control? and they work to capture semantic meaning? and they work to create beautiful art? and they work on creating an AGI chat bot? yeah they'd probably work for proteins too :P | 7 |
| 1624473034005311489 | @GabrielAsher02 @nathanbenaich Not to mention that almost any problem can be framed as a computer vision task. | 7 |
| 1625169361244766213 | @nathanbenaich What happens next is gonna be interesting https://t.co/nCEJkLsHuQ | 6 |
| 1624463386258841603 | @nathanbenaich Good researchers are naturally curious and have broader interests - they will inevitably work on and publish things that are not core to the company they work at imo | 5 |
Top 3 Languages Used In Tweets
Top 10 Hashtags used
| Hashtag | Count |
|---|---|
| #ai | 8 |
| #generativeai | 4 |
| #artificialintelligence | 3 |
| #business | 3 |
| #genai | 3 |
| #stateofai | 2 |
| #fintech | 2 |
| #agi | 2 |
| #machinelearning | 2 |
| #deeplearning | 2 |
Top 10 Hashtags Used In Tweets
Top 10 mentions
| Mention | Count |
|---|---|
| @nathanbenaich | 173 |
| @siftedeu | 6 |
| @shigekzishihara | 4 |
| @strwbilly | 4 |
| @ai_ucl | 4 |
| @ucl_dark | 4 |
| @stateofaireport | 3 |
| @katieelink | 3 |
| @itsalichaudhry | 3 |
| @tkaraletsos | 3 |
Top 10 mentions
Wordcloud of Tweets
Emojis
Average number of emojis used per tweet
20
Emojis used in tweets
| Emoji | Count | Emoji Text |
|---|---|---|
| 👏 | 4 | clapping_hands |
| 👋 | 3 | waving_hand |
| 😂 | 3 | face_with_tears_of_joy |
| 🔥 | 2 | fire |
| 💪 | 1 | flexed_biceps |
| ⚖️ | 1 | balance_scale |
| 🪄 | 1 | magic_wand |
| 😎 | 1 | smiling_face_with_sunglasses |
| 🧵 | 1 | thread |
| 😁 | 1 | beaming_face_with_smiling_eyes |
Emojis groups
| Emoji Group | Count |
|---|---|
| Smileys & Emotion | 13 |
| People & Body | 10 |
| Travel & Places | 4 |
| Activities | 3 |
| Objects | 2 |
| Animals & Nature | 1 |
| Food & Drink | 1 |