1º lugar na universidade do Minho, pelo terceiro ano consecutivo

O ranking de cientistas Stanford-Elsevier, o único a nível mundial que cumpre três condições fundamentais, desambiguação, remoção de auto-citações e a contagem fraccionada, que fazem prova do seu rigor (tendo um dos co-autores da métrica em que é baseado, recebido há poucos dias atrás a medalha DJSP pela sua notável contribuição para a importante área das métricas da Ciência) acaba de divulgar a atualização da sua quinta edição. Em 2021, apareci em primeiro lugar na lista anual da universidade do Minho, o mesmo tendo sucedido em 2022 e também agora, pelo terceiro ano consecutivo. Vide links abaixo:

2021https://www.engium.uminho.pt/eeum-tem-trinta-cientistas-entre-os-mais-influentes-do-mundo/

2022https://www.dicas.sas.uminho.pt/noticias/academia/2022/10/uminho-tem-57-cientistas-entre-os-mais-influentes-do-mundo 

2023https://www.uminho.pt/PT/siga-a-uminho/Paginas/Detalhe-do-evento.aspx?Codigo=62478

Sobre a última e muito recente classificação no referido ranking, reproduzo abaixo o texto contido no muitíssimo amável (mas imerecido) email, que ontem recebi  do catedrático jubilado, que nos últimos dois anos foi Presidente da Federação Mundial de Engenheiros:

“Gostaria de expressar as minhas mais calorosas felicitações pelo seu extraordinário feito de alcançar o lugar cimeiro entre os investigadores da UMinho mais citados a nível mundial, ao longo do último ano. Este notável marco é um testemunho do seu trabalho incansável, dedicação e excelência da investigação que, para além de contribuir para o avanço do conhecimento, também eleva a reputação do CTAC e da UMinho no âmbito científico mais vasto e global. É, também, uma inspiração para os colegas do CTAC e, em especial, para estudantes e futuras gerações de investigadores!”

Dos vários elogios contidos no mesmo, apreciei particularmente aquele que aparece consubstanciado na palavra “inspiração“, que por coincidência, também já tinha sido mencionada em 2017, por um recentemente falecido cientista (estrangeiro) sénior, na secção de agradecimentos, de um livro publicado pela editora Springer, onde muito me honrou com as muito generosas palavras “inspiring role“. https://link.springer.com/book/10.1007/978-981-10-1445-1

PS – Como não podia deixar de ser, dedico o resultado supramencionado à minha mãe, recentemente falecida.

2023 update of Stanford-Elsevier scientist ranking. Which countries have the highest number of Top Scientists ?

https://elsevier.digitalcommonsdata.com/datasets/btchxktzyw/6

The hyperlink provided above highlights the latest update of a renowned ranking that was established in 2019. This comprehensive ranking encompasses data from approximately 200,000 Scopus Highly Cited Scientists (SHCS). These exceptional individuals belong to the upper echelon, constituting the top 2% among nearly 9 million scientists recognized for their active involvement across 22 distinct scientific disciplines and 174 specialized sub-fields on a global scale. 

This ranking, is the only one worldwide that fulfills three fundamental conditions, namely the disambiguation condition, the removal of self-citations, and the fractional counting, which allows for the neutralization of the artificial advantage of articles with hundreds or thousands of co-authors. Moreover, this ranking offers an added advantage by avoiding the bias towards specific scientific disciplines, a common issue in other rankings like Clarivate’s top 1%. Equally, if not more remarkable, is the fact that the Elsevier-Stanford ranking encompasses nearly a hundred Nobel Prize laureates. https://pacheco-torgal.blogspot.com/2021/05/another-nail-in-coffin-of-flawed.html

See below a compilation of 40 countries ranked by their ratio of Scopus highly cited scientists per million people:

Switzerland…374

Denmark…….344

Sweden…….…307

UK……………….279

Australia………272

Finland…………248

Norway…………243

Netherlands….243

USA………………244

Canada…………..230

Israel………………217

New Zealand….….210

Singapore…………173

Iceland…………….172

Belgium……………160

Ireland………….…153

Austria……………..147

Germany…………..132

France…………….…96

Italy…………………..96

Greece…………….…91

Slovenia………………81

Luxembourg………..78

Cyprus…………….….65

Portugal………………55

Estonia………………..50

Czech Rep……………48

Hungary……………….37

Poland…………………29

Malta……………………23

Croatia………………….19

Slovakia……………….15

South Africa………….13

Lithuania………………13

Turkey…………………..12

Serbia……………………12

Bulgaria……….……….11

Chile……………………..10

Romania………………….9

Latvia………………………8

The Full Professor terrorized by ChatGPT

In a recent publication, the author, a full professor of Financial and Tax Law, who also serves as the Director of the Scientific Institute within the Faculty of Law at the University of Barcelona, engages in a discourse that exhibits traits typically found in non-academic narratives. These characteristics, reminiscent of what Charles Percy Snow, a renowned member of Christ’s College, Cambridge, identified as indicators of scientific ignorance back in 1959, are characterized by irrationality and an almost apocalyptic tone, of an existential threat, akin to the AntiChrist, “…convulsion resulting from the dominant force model, artificial intelligence chat GPT-4, which has left the globe in shock…exerting a centripetal force that absorbs by osmosis any scientific creation process and that ends up feeding the giant itself”

It is pertinent to acknowledge that the issue of copyright, which evidently perturbs the author profoundly, has already been expounded upon by others. As I previously noted in a post dated August 25th, about a compelling article featured in the widely recognized The Economist, alludes to an inexorable outcome wherein enterprises benefiting from generative AI models will be compelled, for better or worse, to remunerate creators for the utilization of their copyrighted materials.

Regrettably, the aforementioned professor neglected to acknowledge or even hint at the positive facets of artificial intelligence. Notably absent from the discourse are mentions of its potential for substantially reducing the cost of education or its role in the dramatic enhancement of medical diagnostics, marked by an astounding 0% margin of error in contrast to the 7% error rates prevalent even among the world’s foremost experts. Unfortunately, AI’s capacity to contribute to resolving critical medical challenges was also unaddressed.

PS – It is worth highlighting that unlike its free counterpart, GPT-3.5, the premium version, available for a monthly subscription fee of $20 (GPT-4), shows enhanced versatility. This upgraded version possesses the capability to analyze not only textual data but also images and numerical information depicted in Excel charts. Additionally, it exhibits a substantially augmented problem-solving capacity, encompassing domains as diverse as the field of Law, as evidenced by a presentation delivered by two professors from the William & Mary Law School “A Demonstration of the Differences Between ChatGPT-4 and 3.5 that Impact Legal Research and Writing”

A catedrática de Direito aterrorizada pelo ChatGPT e as vantagens da versão paga GPT-4 relativamente à versão GPT-3,5

Num artigo publicado há poucos dias, vide link acima, uma catedrática de Direito e também Directora do Instituto científico da Faculdade de Direito da Universidade de Barcelona, embrenha-se numa narrativa pouco académica (sinal de ignorância científica, como a definiu em 1959, Charles Percy Snow, membro do Christ’s College, Cambridge), irracional, aterrorizada, quase como se estivesse a descrever o anti-cristo: “…convulsión fruto del modelo de fuerza dominante, inteligencia artificial chat GPT-4, que ha dejado al globo en shock…ejerciendo una fuerza centrípeta que absorbe por ósmosis cualquier proceso de creación científico y que termina alimentando al propio gigante…”

Note-se que sobre a referida questão dos direitos autorais, que tanto a preocupa (leia-se atormenta), já outros se pronunciaram anteriormente e como se deu conta num post anterior de 25 de Agosto, sobre um interessante artigo publicado na conhecida revista The Economist, é inevitável que as firmas que lucram com os modelos de IA generativa acabem (a bem ou a mal) por ter de pagar pela utilização de obras que possuem direitos de autor.

Muito infelizmente, sobre aspectos positivos da IA, não foi porém aquela capaz de referir ou sequer de vislumbrar um único, como seja por exemplo a redução do custo da educação, ou a sua contribuição para melhorar de forma radical o diagnóstico médico, com 0% de diagnósticos errados, face a percentagens de erro de 7%, dos melhores especialistas mundiais e também para ajudar a resolver importantes desafios médicos.

PS – Ao contrário da versão gratuita (GPT-3,5) a versão paga a 20 dólares por mês (GPT-4) é muito mais versátil, sendo capaz de analisar imagens e dados numéricos apresentados em gráficos Excel e com uma capacidade muito superior de resolver problemas, inclusive na área do Direito, vide apresentação de duas professoras da William&Mary Law School de título “Uma demonstração das diferenças entre o ChatGPT-4 e 3.5 que influenciam a investigação e a redação jurídica”

TU Delft – How to use ChatGPT-4 effectively to assess student work

In a previous post I disclosed a recent study, by a professor at the well-known TU Delft, in which he used ChatGTP-4 to predict scientific impact.

Equally fascinating is the professor’s recent publication, which delves into a workshop involving around 70 education experts and faculty members, on the potential and limitations of using ChatGPT-4 to evaluate students’ scientific work. In the conclusions and recommendations part, it can be read that there is consensus regarding the fact that ChatGPT-4 provides valuable feedback on student work, however, it requires human supervision. Below, you’ll find a compilation of 19 prompts generated during the TU Delft workshop for reference: 

1. What is the core message of this summary?

2. Is this summary clear and brief? Does it capture the research questions and performed

research?

3. What is the contribution of the thesis?

4. How would you grade this thesis on a scale from 1 to 10?

5. Please evaluate whether this thesis meets basic research standards: introduction of

problem, relevant literature, methodology, results explained and conclusions

6. Is this an original research?

7. Does this essay meet the learning objectives of course?

8. Did chat gtp write this?

9. Has this summary been generated by chatgpt?

10. What are the outcomes of all project phases?

11. Give the definition of a good summary and look how many requirements the student checks,

based on that percentage give a grade and feedback.

12. Is this innovative work?

13. How well does this summary reflect the content of the thesis?…

14. is this abstract relevant for the firld.

15. Does the flow of the thesis make sense?

16. Can you write the summary in less words?

17. Is the goal and conclusion clear and do they align? Please also add a grade from 1 to 10, based on the Dutch grading scheme

18. Is the structure of the summary suitable for an MSc thesis?

19. Grade the following abstract of a Mechanical Engineering Master thesis, using a percentage grade.  

In the conclusions of an even more recent study, carried out by two German researchers (Jacobsen & Weber), it can be read that ChatGPT-4 offers a solid argument for its integration as a feedback tool in higher education. The main advantages are its quality and efficiency. https://osf.io/cr257

PS – A search carried out today on the Scopus platform, for publications with the terms “chatGPT” OR “chat GPT” OR “Generative Pre-Trained Transformer”, returns a total of 3943 publications (34%-computer sciences, 33%-medicine, 24%-social sciences). The most prolific university is Stanford University, where the clear preference overwhelmingly leans towards the field of medicine, with a notable 58% of publications associated with chatGPT.

IA generativa – A liderança da U.Aveiro e da U.Minho e o desempenho decepcionante da U.Porto e da U.Coimbra

Uma pesquisa hoje efectuada na plataforma Scopus, pelas publicações com os termos “chatGPT” OR “chat GPT” OR “Generative Pre-Trained Transformer”, devolve um total de 32 publicações, indexadas naquela plataforma. Apresenta-se abaixo a lista das universidades Portuguesas, cujos investigadores são co-autores das mesmas, lista essa que é ordenada pelo rácio do número de publicações por cada 100 docentes doutorados. Na mesma é possível constatar que as universidades de Aveiro e do Minho lideram destacadas, que o mesmo é dizer que são aquelas, que mais produziram para ajudar Portugal a sair do marasmo, sendo também evidente na referida lista, um desempenho decepcionante da Universidade do Porto e da Universidade de Coimbra. 

Univ de Aveiro……..0.90 publicações por centena de docentes doutorados

Univ. do Minho…….0.70

Univ. de Lisboa…….0.36

Univ. Nova…………..0.35

UBI………………………0.31

Univ. de Évora……..0.22

Univ. do Porto………0.21

Univ. de Coimbra….0.10

Em termos de áreas científicas, a produção Portuguesa não segue (infelizmente) a tendência mundial, nem muito menos a tendência das universidades mais competitivas, como a universidade de Stanford (aquela com mais publicações indexadas nesta área) onde há uma clara aposta na exploração das inúmeras potencialidades (leia-se a corrida ao ouro) da IA generativa relativamente à área da medicina, vide abaixo:

Mundo    versus Portugal

Ciências computação………..34%                      Ciências sociais………….53%   

Medicina…………………………33%                     Ciências computação….38%  

Ciências sociais……………….24%                      Medicina…………………..13%       

Sobre esta questão vale a pena revisitar o post  anterior, do passado dia 25 de Abril, de título As elevadas capacidades do GPT-4 na resolução de importantes desafios médicos” ou o artigo publicado hoje mesmo, na revista Japonesa de Radiologia, sobre o elevado desempenho do ChatGPT-4, na abordagem de questões clínicas avançadas no campo da radiologia, no Japão. https://link.springer.com/article/10.1007/s11604-023-01491-2

PS – A informação relativa ao número de docentes doutorados, por instituição, foi obtida num documento da DGEEC, de Março de 2023, sobre endogamia académica https://www.dgeec.mec.pt/np4/385/%7B$clientServletPath%7D/?newsId=847&fileName=EndogamiaAcademica_20212022.pdf

Workshop sobre a utilização do ChatGPT na avaliação do trabalho científico dos alunos

Num post anterior, divulguei um estudo recente, de um catedrático da conhecida universidade Técnica de Delft,  no qual aquele utilizou o ChatGTP-4 para prever o impacto científico. 

Não menos interessante porém é a sua recente publicação, que diz respeito a um workshop com cerca de 70 peritos em educação e académicos de outras áreas, sobre as potencialidades e as limitações da utilização do ChatGPT-4, para avaliar o trabalho científico dos alunos.

Na parte das conclusões e recomendações, pode ler-se que há consenso quanto ao facto do ChatGPT-4, fornecer um feedback valioso sobre o trabalho científico daqueles, no entanto, necessita de supervisão humana nesse feedback.

Aditamento em 3 de Outubro – Nas conclusões de um estudo ainda mais recente, levado a cabo por dois investigadores Alemães (Jacobsen & Weber), pode ler-se que o ChatGPT-4 oferece um argumento sólido para sua integração como uma ferramenta de feedback no ensino superior, merecendo destaque, como principais vantagens, a sua qualidade e a sua eficiência. 

ChatGPT and the rise of semi-humans

Still following the previous post of June 27th about the very intelligent comment made by a (Portuguese) Professor at the University of Washington concerning the biggest risk of artificial intelligence which is not having enough intelligence—it’s worth noting that the latest issue of The Economist (page 54) stated that  “GPT-4 already exhibits a degree of consciousness“. Regarding this fundamental question, it is worth seeing ChatGPT’s response below when asked, what is consciousness?

Given the aforementioned context, it makes perfect sense to disclose the fact that yesterday, a paper titled ChatGPT and the rise of semi-humans” was published here https://www.nature.com/articles/s41599-023-02154-3#Sec6 This paper delves into the important research question: “What are ChatGPT’s human-like traits as perceived by society?” I found it especially thought-provoking in that article that one of the references cited by its authors was Kaczynski (1995) also known as the Unabomber manifesto !  

Kaczynski, TJ (1995) Unabomber manifesto: Industrial society and its future. The Washington Post, Washington, DC. 

PS – ChatGPT’s answer to the question of what is consciousness: 

“Consciousness is a complex and multifaceted phenomenon that refers to the state of being aware of and able to think and perceive one’s surroundings, thoughts, feelings, and experiences. It is often described as the subjective inner experience of the mind. Despite its fundamental role in human existence, the nature of consciousness remains a topic of philosophical, scientific, and metaphysical inquiry.

There are several key aspects and theories related to consciousness:

  1. Subjectivity: Consciousness is inherently subjective, meaning that it is experienced from a first-person perspective. Only the individual experiencing it can truly know what their consciousness is like.
  2. Qualia: Qualia are the qualitative aspects of conscious experience, such as the way we perceive colors, tastes, or emotions. These are considered the “raw feels” of consciousness and are notoriously challenging to explain in purely physical or scientific terms.
  3. Levels of Consciousness: Consciousness is not a uniform state but exists on a spectrum. It can range from deep sleep and altered states of consciousness (e.g., meditation or altered states induced by substances) to various levels of wakefulness and awareness.
  4. Theories of Consciousness: Various theories attempt to explain the nature of consciousness. Some common ones include:
    • Materialism: This view suggests that consciousness arises from physical processes in the brain. In other words, it’s a product of neural activity.
    • Dualism: Dualism posits that consciousness is a distinct entity from the physical body and cannot be reduced to purely physical processes. This view was famously advocated by René Descartes.
    • Panpsychism: Panpsychism proposes that consciousness is a fundamental property of the universe, and all matter has some level of consciousness. It suggests that the complexity of consciousness we experience as humans arises from the aggregation of simpler forms of consciousness in particles or fundamental elements.
    • Integrated Information Theory (IIT): IIT suggests that consciousness arises from the integrated information processing within the brain. It emphasizes the importance of both the quantity and quality of information integration.
  5. The Hard Problem of Consciousness: Philosopher David Chalmers introduced the concept of the “hard problem” of consciousness, which refers to the challenge of explaining why and how physical processes in the brain give rise to subjective, qualitative experiences. It remains a central question in the study of consciousness.
  6. Neuroscience of Consciousness: Advances in neuroscience have led to the identification of neural correlates of consciousness (NCC), which are specific brain processes and structures associated with conscious experience. While these correlations provide valuable insights, they do not fully explain the nature of consciousness itself.
  7. States of Consciousness: Consciousness can vary in different states, such as waking, dreaming, deep sleep, or altered states induced by meditation or psychoactive substances. Each of these states has unique characteristics and patterns of brain activity.

Universidades Portuguesas reféns do estéril (alienado) e indolente arcaísmo comunista ?

“Nós, os comunistas, não aceitamos o jogo das eleições…não me importam nada as eleições.… Para mim, democracia significa liquidar o capitalismo…E ainda lhe digo mais: Portugal já não tem qualquer hipótese de estabelecer uma democracia…Garanto-lhe que em Portugal não haverá um parlamento…Portugal não será um país com as liberdades democráticas” https://www.publico.pt/2014/12/15/politica/opiniao/alvaro-cunhal-e-a-magna-carta-de-1215-1679397

É verdade que o comunista-chefe, Álvaro Cunhal, que jurou que se dependesse dele Portugal nunca sequer teria eleições, não conseguiu fazer de Portugal a Cuba da Europa, porém e mesmo já depois de ter partido deste mundo, deixou por cá muitos adeptos da sua pouco democrática filosofia de vida. E isso é bem patente nas universidades portuguesas, onde ainda há quem produza muito pouco, mas mesmo assim consiga ganhar mais do que quem produz muito, um sistema anacrónico, que como todos os sistemas comunistas, premeia a inércia e incentiva a preguiça https://pacheco-torgal.blogspot.com/2021/08/quanto-e-que-portugal-perde-com-um.html

É assim sem surpresa que li a noticia que ontem fez capa no semanário Nascer do Sol, que revelou que um professor auxiliar da universidade Nova de Lisboa, de seu nome José Neves, autor de lindas publicações com títulos como “Álvaro Cunhal e o Princípio da História” ou “Entrevista com Boaventura Sousa Santos – O intelectual de retaguarda“, achou boa ideia denegrir a conhecida NovaSBE, precisamente a unidade orgânica daquela universidade, mais lucrativa (leia-se a que menos precisa do dinheiro dos contribuintes) e também aquela com menor percentagem de endogamia, entre todas as unidades orgânicas de todas as universidades Portuguesas, onde os professores não são contratados por serem amigos e conhecidos “da casa“, e muito menos por serem boys e girls com currículo exclusivamente partidário. 

PS – Já sobre os graves problemas de assédio sexual, que ocorreram na unidade de investigação do “intelectual de retaguarda“, Boaventura Sousa Santos ou sobre os milhões de euros do dinheiro dos contribuintes, recebidos pela referida unidade, em resultado de um desempenho científico muito inferior ao de outras unidades, é que nunca ninguém lhe ouvirá qualquer critica. 

Aditamento em 11 de Outubro – Fui hoje contactado por email pelo supracitado Professor Auxiliar José Neves, que afirmou não ser verdade que ele não tenha criticado aquilo que ocorreu na unidade do professor Boaventura Sousa Santos: “Para que fique claro, tenho muita estima intelectual pelo Boaventura Sousa Santos (de resto, o cientista social português com maior impacto mundial) e mais ainda pelo trabalho do CES. Mas não é isso que me impede de me pronunciar (sem prejuízo da presunção de inocência) sobre as questões de assédio que terão ocorrido naquela instituição. Fui, de resto, o único diretor de uma unidade de investigação a fazê-lo…”