Questões do ENEM 2017

Questões aplicadas na prova de 2017, cada uma em página própria com enunciado, alternativas e gabarito.

Resultados

1.793 questões encontradas. Mostrando a página 47 de 90.

  • 6B741CFB-B5

    Português

    Interpretação de Textos
    Universidade Federal de Juiz de Fora - MG · 2017DifícilEntre para guardar nos favoritos

    Figura 1 de 2 da questão de Português, Universidade Federal de Juiz de Fora - MG 2017, Interpretação de Textos

    Figura 2 de 2 da questão de Português, Universidade Federal de Juiz de Fora - MG 2017, Interpretação de Textos

    Releia a frase:

    Há também gírias características de uma geração, muitas vezes empregadas por adolescentes como uma forma de se distinguir dos mais jovens e dos mais velhos. Como observa o sociolinguista Anthony Julius Naro, gírias adquiridas nesse período continuam sendo empregadas ao longo da vida, como marca da juventude.


    Segundo Naro, é correto afirmar que a gíria:

    Escolha uma alternativa para a questão 6b741cfb-b5
  • 6B6F1A3F-B5

    Português

    Interpretação de Textos
    Universidade Federal de Juiz de Fora - MG · 2017FácilEntre para guardar nos favoritos

    Figura 1 de 2 da questão de Português, Universidade Federal de Juiz de Fora - MG 2017, Interpretação de Textos

    Figura 2 de 2 da questão de Português, Universidade Federal de Juiz de Fora - MG 2017, Interpretação de Textos

    De acordo com o texto, é correto afirmar que as gírias são
    Escolha uma alternativa para a questão 6b6f1a3f-b5
  • 5F817E9C-B5

    Inglês

    Interpretação de texto | Reading comprehension
    IF-PR · 2017DifícilEntre para guardar nos favoritos

    August 16, 2017 / 10:00 AM

    Four years without Big Ben’s bongs? It can’t be right, says UK PM May

        LONDON (Reuters) - Prime Minister Theresa May said on Wednesday it could not be right for “Big Ben”, the bell in the British parliament’s clock tower whose bongs (1)_______, to fall silent for four years during renovations. May joined other politicians who have protested at the news that the great bell, which has rung every hour for most of the past 157 years, would cease its bongs to ensure the safety of workers carrying out renovations on the tower.

        “Of course we want to ensure that people are safe at work, but it can’t be right for Big Ben to be silent for four years,” May told reporters.

        “I hope that the Speaker (of the House of Commons) ... will urgently look into this and ensure that we can hear Big Ben through those four years.”

        Big Ben’s bongs, which are heard marking the start of some of the BBC’s flagship news bulletins, are part of the soundtrack of daily life in the British capital and beyond.

        Officially known as the Elizabeth Tower, the clock tower that houses Big Ben is believed to be the most photographed building in the United Kingdom. May’s comments were more restrained than those of her Brexit minister, David Davis, who said on Tuesday that (2)_____.

        Another Conservative politician, member of parliament Nigel Evans, suggested earlier on Wednesday that the bongs could be switched back on every evening when the workers carrying out the renovations of the clock tower finished for the day.

        Steve Jaggs, parliament’s Keeper of the Great Clock, had announced this week that Big Ben would stop its regular chimes at midday (1100 GMT) on Monday, Aug. 21, inviting members of the public to gather nearby to hear the final bongs.

        The bell will still toll for important (3)_____ but will otherwise remain silent until 2021.

    Reporting by Estelle Shirbon; editing by Michael Holden (Adaptado de Acesso em 16/08/2017) 

    Which” in the sentence “(…) the great bell, which has rung every hour for most of the past 157 years, would cease its bongs (…)” refers to:
    Escolha uma alternativa para a questão 5f817e9c-b5
  • 5F7D6057-B5

    Inglês

    Interpretação de texto | Reading comprehension
    IF-PR · 2017MédioEntre para guardar nos favoritos

    August 16, 2017 / 10:00 AM

    Four years without Big Ben’s bongs? It can’t be right, says UK PM May

        LONDON (Reuters) - Prime Minister Theresa May said on Wednesday it could not be right for “Big Ben”, the bell in the British parliament’s clock tower whose bongs (1)_______, to fall silent for four years during renovations. May joined other politicians who have protested at the news that the great bell, which has rung every hour for most of the past 157 years, would cease its bongs to ensure the safety of workers carrying out renovations on the tower.

        “Of course we want to ensure that people are safe at work, but it can’t be right for Big Ben to be silent for four years,” May told reporters.

        “I hope that the Speaker (of the House of Commons) ... will urgently look into this and ensure that we can hear Big Ben through those four years.”

        Big Ben’s bongs, which are heard marking the start of some of the BBC’s flagship news bulletins, are part of the soundtrack of daily life in the British capital and beyond.

        Officially known as the Elizabeth Tower, the clock tower that houses Big Ben is believed to be the most photographed building in the United Kingdom. May’s comments were more restrained than those of her Brexit minister, David Davis, who said on Tuesday that (2)_____.

        Another Conservative politician, member of parliament Nigel Evans, suggested earlier on Wednesday that the bongs could be switched back on every evening when the workers carrying out the renovations of the clock tower finished for the day.

        Steve Jaggs, parliament’s Keeper of the Great Clock, had announced this week that Big Ben would stop its regular chimes at midday (1100 GMT) on Monday, Aug. 21, inviting members of the public to gather nearby to hear the final bongs.

        The bell will still toll for important (3)_____ but will otherwise remain silent until 2021.

    Reporting by Estelle Shirbon; editing by Michael Holden (Adaptado de Acesso em 16/08/2017) 

    When will the bongs begin again regularly?
    Escolha uma alternativa para a questão 5f7d6057-b5
  • 5F797898-B5

    Inglês

    Interpretação de texto | Reading comprehension
    IF-PR · 2017Muito difícilEntre para guardar nos favoritos

    August 16, 2017 / 10:00 AM

    Four years without Big Ben’s bongs? It can’t be right, says UK PM May

        LONDON (Reuters) - Prime Minister Theresa May said on Wednesday it could not be right for “Big Ben”, the bell in the British parliament’s clock tower whose bongs (1)_______, to fall silent for four years during renovations. May joined other politicians who have protested at the news that the great bell, which has rung every hour for most of the past 157 years, would cease its bongs to ensure the safety of workers carrying out renovations on the tower.

        “Of course we want to ensure that people are safe at work, but it can’t be right for Big Ben to be silent for four years,” May told reporters.

        “I hope that the Speaker (of the House of Commons) ... will urgently look into this and ensure that we can hear Big Ben through those four years.”

        Big Ben’s bongs, which are heard marking the start of some of the BBC’s flagship news bulletins, are part of the soundtrack of daily life in the British capital and beyond.

        Officially known as the Elizabeth Tower, the clock tower that houses Big Ben is believed to be the most photographed building in the United Kingdom. May’s comments were more restrained than those of her Brexit minister, David Davis, who said on Tuesday that (2)_____.

        Another Conservative politician, member of parliament Nigel Evans, suggested earlier on Wednesday that the bongs could be switched back on every evening when the workers carrying out the renovations of the clock tower finished for the day.

        Steve Jaggs, parliament’s Keeper of the Great Clock, had announced this week that Big Ben would stop its regular chimes at midday (1100 GMT) on Monday, Aug. 21, inviting members of the public to gather nearby to hear the final bongs.

        The bell will still toll for important (3)_____ but will otherwise remain silent until 2021.

    Reporting by Estelle Shirbon; editing by Michael Holden (Adaptado de Acesso em 16/08/2017) 

    What is the official name of the clock tower that houses the famous bell?
    Escolha uma alternativa para a questão 5f797898-b5
  • 5F764872-B5

    Inglês

    Interpretação de texto | Reading comprehension
    IF-PR · 2017DifícilEntre para guardar nos favoritos

    August 16, 2017 / 10:00 AM

    Four years without Big Ben’s bongs? It can’t be right, says UK PM May

        LONDON (Reuters) - Prime Minister Theresa May said on Wednesday it could not be right for “Big Ben”, the bell in the British parliament’s clock tower whose bongs (1)_______, to fall silent for four years during renovations. May joined other politicians who have protested at the news that the great bell, which has rung every hour for most of the past 157 years, would cease its bongs to ensure the safety of workers carrying out renovations on the tower.

        “Of course we want to ensure that people are safe at work, but it can’t be right for Big Ben to be silent for four years,” May told reporters.

        “I hope that the Speaker (of the House of Commons) ... will urgently look into this and ensure that we can hear Big Ben through those four years.”

        Big Ben’s bongs, which are heard marking the start of some of the BBC’s flagship news bulletins, are part of the soundtrack of daily life in the British capital and beyond.

        Officially known as the Elizabeth Tower, the clock tower that houses Big Ben is believed to be the most photographed building in the United Kingdom. May’s comments were more restrained than those of her Brexit minister, David Davis, who said on Tuesday that (2)_____.

        Another Conservative politician, member of parliament Nigel Evans, suggested earlier on Wednesday that the bongs could be switched back on every evening when the workers carrying out the renovations of the clock tower finished for the day.

        Steve Jaggs, parliament’s Keeper of the Great Clock, had announced this week that Big Ben would stop its regular chimes at midday (1100 GMT) on Monday, Aug. 21, inviting members of the public to gather nearby to hear the final bongs.

        The bell will still toll for important (3)_____ but will otherwise remain silent until 2021.

    Reporting by Estelle Shirbon; editing by Michael Holden (Adaptado de Acesso em 16/08/2017) 

    Why will Big Ben be silent for four years?
    Escolha uma alternativa para a questão 5f764872-b5
  • 5F71D492-B5

    Inglês

    Interpretação de texto | Reading comprehension
    IF-PR · 2017DifícilEntre para guardar nos favoritos

    August 16, 2017 / 10:00 AM

    Four years without Big Ben’s bongs? It can’t be right, says UK PM May

        LONDON (Reuters) - Prime Minister Theresa May said on Wednesday it could not be right for “Big Ben”, the bell in the British parliament’s clock tower whose bongs (1)_______, to fall silent for four years during renovations. May joined other politicians who have protested at the news that the great bell, which has rung every hour for most of the past 157 years, would cease its bongs to ensure the safety of workers carrying out renovations on the tower.

        “Of course we want to ensure that people are safe at work, but it can’t be right for Big Ben to be silent for four years,” May told reporters.

        “I hope that the Speaker (of the House of Commons) ... will urgently look into this and ensure that we can hear Big Ben through those four years.”

        Big Ben’s bongs, which are heard marking the start of some of the BBC’s flagship news bulletins, are part of the soundtrack of daily life in the British capital and beyond.

        Officially known as the Elizabeth Tower, the clock tower that houses Big Ben is believed to be the most photographed building in the United Kingdom. May’s comments were more restrained than those of her Brexit minister, David Davis, who said on Tuesday that (2)_____.

        Another Conservative politician, member of parliament Nigel Evans, suggested earlier on Wednesday that the bongs could be switched back on every evening when the workers carrying out the renovations of the clock tower finished for the day.

        Steve Jaggs, parliament’s Keeper of the Great Clock, had announced this week that Big Ben would stop its regular chimes at midday (1100 GMT) on Monday, Aug. 21, inviting members of the public to gather nearby to hear the final bongs.

        The bell will still toll for important (3)_____ but will otherwise remain silent until 2021.

    Reporting by Estelle Shirbon; editing by Michael Holden (Adaptado de Acesso em 16/08/2017) 

    In the previous text, three parts of sentences have been removed. Below you will find the three removed parts PLUS one which doesn’t fit. Choose from the parts of sentences (I – IV) the one which fits each gap (1 – 3). Remember, there is one extra sentence you do not need to use.
    Sentences
    I) the House of Commons said on Monday
    II) events such as New Year’s Eve celebrations
    III) are one of the country’s most familiar sounds
    IV) the silencing of Big Ben for such a long period was “mad”

    Now choose the correct alternative.
    Escolha uma alternativa para a questão 5f71d492-b5
  • 5F6C5EAC-B5

    Português

    Interpretação de Textos
    IF-PR · 2017DifícilEntre para guardar nos favoritos

    O texto a seguir servirá de base para a questão.

        Os financiamentos imobiliários com recursos da poupança ([Sistema Brasileiro de Poupança e Empréstimo] SBPE) totalizaram R$ 20,6 bilhões no primeiro semestre deste ano, montante 9,1% inferior ao registrado em igual intervalo de 2016, de acordo com a Associação Brasileira das Entidades de Crédito Imobiliário e Poupança (Abecip).

        Foram financiadas, com recursos das cadernetas de poupança, 82,5 mil unidades entre aquisições e construções no período, recuo de 17,9% em relação à primeira metade de 2016, quando 100,5 mil imóveis foram financiados.

        Apenas em junho, conforme a Abecip, o montante de financiamento imobiliário alcançou R$ 3,8 bilhões, alta de 6,5% em relação a maio, mas retração de 11,1% no período de um ano. A quantidade de imóveis financiados no mês passado subiu 5,7% e recuou 22%, nas mesmas bases de comparações, para 15,4 mil unidades.

        “O primeiro semestre foi ruim. Esperamos decrescimento menor ou ao menos empate com o ano passado e isso não aconteceu, mas o patamar de empréstimos continuou interessante”, disse o presidente da Abecip, Gilberto Abreu, em coletiva de imprensa, lembrando que as empresas estão em um ritmo menor de construção de novos empreendimentos após um ciclo de elevados estoques (Estadão, 26/07/2017).

    Releia o trecho negritado no texto e transcrito a seguir: “A quantidade de imóveis financiados no mês passado subiu 5,7% e recuou 22%, nas mesmas bases de comparações, para 15,4 mil unidades”. O significado de “nas mesmas bases” está explicitado na alternativa:
    Escolha uma alternativa para a questão 5f6c5eac-b5
  • 5F686A3B-B5

    Português

    Conjunções: Relação de causa e consequência
    IF-PR · 2017MédioEntre para guardar nos favoritos

    O texto a seguir servirá de base para a questão.

        Os financiamentos imobiliários com recursos da poupança ([Sistema Brasileiro de Poupança e Empréstimo] SBPE) totalizaram R$ 20,6 bilhões no primeiro semestre deste ano, montante 9,1% inferior ao registrado em igual intervalo de 2016, de acordo com a Associação Brasileira das Entidades de Crédito Imobiliário e Poupança (Abecip).

        Foram financiadas, com recursos das cadernetas de poupança, 82,5 mil unidades entre aquisições e construções no período, recuo de 17,9% em relação à primeira metade de 2016, quando 100,5 mil imóveis foram financiados.

        Apenas em junho, conforme a Abecip, o montante de financiamento imobiliário alcançou R$ 3,8 bilhões, alta de 6,5% em relação a maio, mas retração de 11,1% no período de um ano. A quantidade de imóveis financiados no mês passado subiu 5,7% e recuou 22%, nas mesmas bases de comparações, para 15,4 mil unidades.

        “O primeiro semestre foi ruim. Esperamos decrescimento menor ou ao menos empate com o ano passado e isso não aconteceu, mas o patamar de empréstimos continuou interessante”, disse o presidente da Abecip, Gilberto Abreu, em coletiva de imprensa, lembrando que as empresas estão em um ritmo menor de construção de novos empreendimentos após um ciclo de elevados estoques (Estadão, 26/07/2017).

    Analise o último parágrafo do texto, transcrito a seguir, e depois escolha a alternativa correta.
    O primeiro semestre foi ruim. Esperamos (1) decrescimento (2) menor ou ao menos empate (3) com o ano passado e (4) isso não aconteceu, mas o patamar de empréstimos continuou interessante”, disse o presidente da Abecip, Gilberto Abreu, em coletiva de imprensa, lembrando que as empresas estão em um ritmo menor de construção de novos empreendimentos após um ciclo de elevados estoques.
    Há uma inadequação:
    Escolha uma alternativa para a questão 5f686a3b-b5
  • 5F651E87-B5

    Português

    Interpretação de Textos
    IF-PR · 2017MédioEntre para guardar nos favoritos

    O texto a seguir servirá de base para a questão.

        Os financiamentos imobiliários com recursos da poupança ([Sistema Brasileiro de Poupança e Empréstimo] SBPE) totalizaram R$ 20,6 bilhões no primeiro semestre deste ano, montante 9,1% inferior ao registrado em igual intervalo de 2016, de acordo com a Associação Brasileira das Entidades de Crédito Imobiliário e Poupança (Abecip).

        Foram financiadas, com recursos das cadernetas de poupança, 82,5 mil unidades entre aquisições e construções no período, recuo de 17,9% em relação à primeira metade de 2016, quando 100,5 mil imóveis foram financiados.

        Apenas em junho, conforme a Abecip, o montante de financiamento imobiliário alcançou R$ 3,8 bilhões, alta de 6,5% em relação a maio, mas retração de 11,1% no período de um ano. A quantidade de imóveis financiados no mês passado subiu 5,7% e recuou 22%, nas mesmas bases de comparações, para 15,4 mil unidades.

        “O primeiro semestre foi ruim. Esperamos decrescimento menor ou ao menos empate com o ano passado e isso não aconteceu, mas o patamar de empréstimos continuou interessante”, disse o presidente da Abecip, Gilberto Abreu, em coletiva de imprensa, lembrando que as empresas estão em um ritmo menor de construção de novos empreendimentos após um ciclo de elevados estoques (Estadão, 26/07/2017).

    Estratégias argumentativas são mecanismos utilizados em um texto, com o objetivo de convencer o leitor a respeito do assunto que desenvolve. Elas podem ser de diversos tipos, entre os quais:
    I) apresentação de argumentos concretos, como dados estatísticos, por exemplo;
    II) utilização de raciocínio lógico, que apresente comparação, por exemplo;
    III) exemplificação, relacionada ao assunto que se desenvolve;
    IV) alusão histórica, em que se busca a intertextualidade;
    V) argumento de autoridade, com a opinião de especialistas, para dar credibilidade ao texto.

    Indique, pela numeração acima, as estratégias presentes no artigo do Estadão:
    Escolha uma alternativa para a questão 5f651e87-b5
  • 5F62327F-B5

    Português

    Adjetivos
    IF-PR · 2017DifícilEntre para guardar nos favoritos

    Imagem da questão de Português, IF-PR 2017, Adjetivos

    Assinale a alternativa que condiz com a articulação entre texto e imagem nos dois anúncios.
    Escolha uma alternativa para a questão 5f62327f-b5
  • 5F5F378B-B5

    Português

    Denotação e Conotação
    IF-PR · 2017DifícilEntre para guardar nos favoritos

    Imagem da questão de Português, IF-PR 2017, Denotação e Conotação

    A respeito dos anúncios, analise as afirmativas a seguir.

    I) Ambos os textos exploram o recurso da polissemia.

    II) Os textos são impróprios; utilizam palavras ofensivas ou da gíria.

    III) O primeiro texto utiliza o recurso da intertextualidade, remetendo a “pai dos burros”.

    IV) No segundo texto, “pública” e “privada” são antônimos.

    V) As palavras “burro” e “privada” foram usadas em seu sentido denotativo.

    Estão corretas apenas:

    Escolha uma alternativa para a questão 5f5f378b-b5
  • 5F5BFB28-B5

    Português

    Interpretação de Textos
    IF-PR · 2017MédioEntre para guardar nos favoritos

    Leia o poema Açúcar, de Ferreira Gullar, com atenção.


    Em lugares distantes, onde não há hospital

    nem escola homens que não sabem ler e morrem de fome

    aos 27 anos

    plantaram e colheram a cana

    que viraria açúcar.


    Em usinas escuras,

    homens de vida amarga

    e dura

    produziram este açúcar

    branco e puro

    com que adoço meu café esta manhã em Ipanema.



    Os recursos expressivos e o enfoque do tema, oferecem ao texto um caráter:

    Escolha uma alternativa para a questão 5f5bfb28-b5
  • 5F569655-B5

    Português

    Interpretação de Textos
    IF-PR · 2017MédioEntre para guardar nos favoritos

    Leia o texto a seguir.

    E se o oceano incendiar

    E se cair neve no sertão

    E se o urubu cocorocar

    E se o botafogo for campeão

    E se o meu dinheiro não faltar

    E se o delegado for gentil

    E se tiver bife no jantar

    E se o carnaval cair em abril

    E se o telefone funcionar

    (...)

    E se tiver sopa pro peão (Francis Hime)


    Assinale a alternativa correta.

    Escolha uma alternativa para a questão 5f569655-b5
  • 5F520589-B5

    Português

    Coesão e coerência
    IF-PR · 2017DifícilEntre para guardar nos favoritos
    Assinale a alternativa em que o texto está redigido corretamente, do ponto de vista da coesão textual.
    Escolha uma alternativa para a questão 5f520589-b5
  • 5F4F139C-B5

    Português

    Interpretação de Textos
    IF-PR · 2017MédioEntre para guardar nos favoritos

    Imagem da questão de Português, IF-PR 2017, Interpretação de Textos

    Sobre a tirinha, é possível afirmar que:
    Escolha uma alternativa para a questão 5f4f139c-b5
  • 5F48DFD0-B5

    Português

    Interpretação de Textos
    IF-PR · 2017MédioEntre para guardar nos favoritos

    Imagem da questão de Português, IF-PR 2017, Interpretação de Textos

    Assinale a alternativa que melhor registra a mensagem que a tirinha pretende veicular.
    Escolha uma alternativa para a questão 5f48dfd0-b5
  • 3C754622-B5

    Inglês

    Interpretação de texto | Reading comprehension
    IF Sul - MG · 2017Muito difícilEntre para guardar nos favoritos
    AI Picks Up Racial and Gender Biases When Learning from What Humans Write

    AI1 picks up racial and gender biases2 when learning language from text, researchers say. Without any supervision, a machine learning algorithm learns to associate female names more with family words than career words, and black names as being more unpleasant than white names.

    For a study published today in Science, researchers tested the bias of a common AI model, and then matched the results against a well-known psychological test that measures bias in humans. The team replicated in the algorithm all the psychological biases they tested, according to a study from co-author Aylin Caliskan, a post-doc at Princeton University. Because machine learning algorithms are so common, influencing everything from translation to scanning names on resumes, this research shows that the biases are pervasive, too. 

    An algorithm is a set of instructions that humans write to help computers learn. Think of it like a recipe, says Zachary Lipton, an AI researcher at UC San Diego who was not involved in the study. Because algorithms use existing materials — like books or text on the internet — it’s obvious that AI can pick up biases if the materials themselves are biased. (For example, Google Photos tagged black users as gorillas.) We’ve known for a while, for instance, that language algorithms learn to associate the word “man” with “professor” and the word “woman” with “assistant professor.” But this paper is interesting because it incorporates previous work done in psychology on human biases, Lipton says.

    For today’s study, Caliskan’s team created a test that resembles the Implicit Association Test (IAT), which is commonly used in psychology to measure how biased people are (though there has been some controversy over its accuracy). In the IAT, subjects are presented with two images — say, a white man and a black man — and words like “pleasant” or “unpleasant.” The IAT calculates how quickly you match up “white man” and “pleasant” versus “black man” and “pleasant,” and vice versa. The idea is that the longer it takes you to match up two concepts, the more trouble you have associating them.

    The test developed by the researchers also calculates bias, but instead of measuring “response time”, it measures the mathematical distance between two words. In other words, if there’s a bigger numerical distance between a black name and the concept of “pleasant” than a white name and “pleasant”, the model’s association between the two isn’t as strong. The further apart the words are, the less the algorithm associates them together.

    Caliskan’s team then tested their method on one particular algorithm: Global Vectors for Word Representation (GLoVe) from Stanford University. GLoVe basically crawls the web to find data and learns associations between billions of words. The researchers found that, in GLoVe, female words are more associated with arts than with math or science, and black names are seen as more unpleasant than white names. That doesn’t mean there’s anything wrong with the AI system, per se, or how the AI is learning — there’s something wrong with the material.

    1AI: Artificial Intelligence
    2bias: prejudice; preconception

    Disponível em <http://www.theverge.com/>. Acesso em: 18/04/2017.
    Why does Artificial Intelligence (AI) pick up biases?
    Escolha uma alternativa para a questão 3c754622-b5
  • 3C728872-B5

    Inglês

    Interpretação de texto | Reading comprehension
    IF Sul - MG · 2017DifícilEntre para guardar nos favoritos
    AI Picks Up Racial and Gender Biases When Learning from What Humans Write

    AI1 picks up racial and gender biases2 when learning language from text, researchers say. Without any supervision, a machine learning algorithm learns to associate female names more with family words than career words, and black names as being more unpleasant than white names.

    For a study published today in Science, researchers tested the bias of a common AI model, and then matched the results against a well-known psychological test that measures bias in humans. The team replicated in the algorithm all the psychological biases they tested, according to a study from co-author Aylin Caliskan, a post-doc at Princeton University. Because machine learning algorithms are so common, influencing everything from translation to scanning names on resumes, this research shows that the biases are pervasive, too. 

    An algorithm is a set of instructions that humans write to help computers learn. Think of it like a recipe, says Zachary Lipton, an AI researcher at UC San Diego who was not involved in the study. Because algorithms use existing materials — like books or text on the internet — it’s obvious that AI can pick up biases if the materials themselves are biased. (For example, Google Photos tagged black users as gorillas.) We’ve known for a while, for instance, that language algorithms learn to associate the word “man” with “professor” and the word “woman” with “assistant professor.” But this paper is interesting because it incorporates previous work done in psychology on human biases, Lipton says.

    For today’s study, Caliskan’s team created a test that resembles the Implicit Association Test (IAT), which is commonly used in psychology to measure how biased people are (though there has been some controversy over its accuracy). In the IAT, subjects are presented with two images — say, a white man and a black man — and words like “pleasant” or “unpleasant.” The IAT calculates how quickly you match up “white man” and “pleasant” versus “black man” and “pleasant,” and vice versa. The idea is that the longer it takes you to match up two concepts, the more trouble you have associating them.

    The test developed by the researchers also calculates bias, but instead of measuring “response time”, it measures the mathematical distance between two words. In other words, if there’s a bigger numerical distance between a black name and the concept of “pleasant” than a white name and “pleasant”, the model’s association between the two isn’t as strong. The further apart the words are, the less the algorithm associates them together.

    Caliskan’s team then tested their method on one particular algorithm: Global Vectors for Word Representation (GLoVe) from Stanford University. GLoVe basically crawls the web to find data and learns associations between billions of words. The researchers found that, in GLoVe, female words are more associated with arts than with math or science, and black names are seen as more unpleasant than white names. That doesn’t mean there’s anything wrong with the AI system, per se, or how the AI is learning — there’s something wrong with the material.

    1AI: Artificial Intelligence
    2bias: prejudice; preconception

    Disponível em <http://www.theverge.com/>. Acesso em: 18/04/2017.
    Com relação ao teste desenvolvido pelos pesquisadores para calcular o preconceito, assinale a alternativa correta.
    Escolha uma alternativa para a questão 3c728872-b5
  • 3C6DFA34-B5

    Inglês

    Interpretação de texto | Reading comprehension
    IF Sul - MG · 2017MédioEntre para guardar nos favoritos
    AI Picks Up Racial and Gender Biases When Learning from What Humans Write

    AI1 picks up racial and gender biases2 when learning language from text, researchers say. Without any supervision, a machine learning algorithm learns to associate female names more with family words than career words, and black names as being more unpleasant than white names.

    For a study published today in Science, researchers tested the bias of a common AI model, and then matched the results against a well-known psychological test that measures bias in humans. The team replicated in the algorithm all the psychological biases they tested, according to a study from co-author Aylin Caliskan, a post-doc at Princeton University. Because machine learning algorithms are so common, influencing everything from translation to scanning names on resumes, this research shows that the biases are pervasive, too. 

    An algorithm is a set of instructions that humans write to help computers learn. Think of it like a recipe, says Zachary Lipton, an AI researcher at UC San Diego who was not involved in the study. Because algorithms use existing materials — like books or text on the internet — it’s obvious that AI can pick up biases if the materials themselves are biased. (For example, Google Photos tagged black users as gorillas.) We’ve known for a while, for instance, that language algorithms learn to associate the word “man” with “professor” and the word “woman” with “assistant professor.” But this paper is interesting because it incorporates previous work done in psychology on human biases, Lipton says.

    For today’s study, Caliskan’s team created a test that resembles the Implicit Association Test (IAT), which is commonly used in psychology to measure how biased people are (though there has been some controversy over its accuracy). In the IAT, subjects are presented with two images — say, a white man and a black man — and words like “pleasant” or “unpleasant.” The IAT calculates how quickly you match up “white man” and “pleasant” versus “black man” and “pleasant,” and vice versa. The idea is that the longer it takes you to match up two concepts, the more trouble you have associating them.

    The test developed by the researchers also calculates bias, but instead of measuring “response time”, it measures the mathematical distance between two words. In other words, if there’s a bigger numerical distance between a black name and the concept of “pleasant” than a white name and “pleasant”, the model’s association between the two isn’t as strong. The further apart the words are, the less the algorithm associates them together.

    Caliskan’s team then tested their method on one particular algorithm: Global Vectors for Word Representation (GLoVe) from Stanford University. GLoVe basically crawls the web to find data and learns associations between billions of words. The researchers found that, in GLoVe, female words are more associated with arts than with math or science, and black names are seen as more unpleasant than white names. That doesn’t mean there’s anything wrong with the AI system, per se, or how the AI is learning — there’s something wrong with the material.

    1AI: Artificial Intelligence
    2bias: prejudice; preconception

    Disponível em <http://www.theverge.com/>. Acesso em: 18/04/2017.
    Assinale a alternativa que está de acordo com o texto.
    Escolha uma alternativa para a questão 3c6dfa34-b5