Go to main contentGo to main menu

Logo de la UAMUniversidad Autónoma de Madrid

Noticias

Another step towards popularising the Semantic Web

05/07/2010

Researchers from the Universidad Autónoma de Madrid (UAM) and the Polytechnic University (UPM) have developed Fortunata, a set  of computer tools that allow us to create semantic web applications in a simple way.

It will soon be the 10th anniversary of the publication of the first article to describe the Semantic Web. This article signed by Tim Berners-Lee, the inventor of the Web, among others, described a new Web in which the applications were able to understand our wishes, expressed in a natural language, and provide us with the information requested in a fast and efficient way. Since then, several international standards have been developed; true technological pillars on which to base the Web of the future.

The current Web contains billions of documents in different formats. In order to search for the information contained in these documents we have search engines, such as Google, with response times of tenths of a second. But search engines have limitations. One example: if you want to search for “books by García Márquez”, simply entering this text in your favourite search engine will lead you to different documents with the desired information. However, if you wanted to know “what books mention one of García Márquez's books?”, the response would be very different… and erroneous. Why? Because search engines do not understand the semantics of the question. In fact, not even the programmes that index Web documents “understand” the semantics of these documents.

Current searches are based on “key words”. These programmes scrutinise the web in search of new documents, but they do not try to “understand” them, they simple extract their words; although it is true that they are very efficient.

Even though there are already standards to provide this data with semantics, and recently the way of consulting this semantic data has been standardised, it is still very complicated to create web applications that use and generate semantic data.

The work of the researchers at the Universidad Autónoma de Madrid (UAM) and the Polytechnic University (UPM) has created a set of computer tools that make the development of this type of application easier. With these tools, a developer who does not have knowledge of Semantic Web technologies can create web applications that use and generate semantic data. Experimental results show that the web applications generated with this infrastructure have a high level of usability and are satisfactory to users, regardless of their level of computing knowledge.

The benefits of this type of application are not immediate, but nor were those of the first websites. A certain critical mass must be achieved that will compensate the technological investment. Currently, the volume of semantic data on the Internet is much smaller than the volume of documents (non-semantic), but its quality is undoubtedly greater, and this type of application contributes to popularising the use and generation of semantic data. It is worth highlighting that the web applications developed with this infrastructure do not change in appearance or functionality in relation to traditional web applications, so that the user of these applications is not conscious that they are using or generating semantic information.

Although great progress has been made, there is still a lot of work to be done before we can enjoy the world-wide semantic web that the father of the Web outlined. Aspects such as natural language processing or interaction with semantic data are highly complex and have not yet left the research labs. A bit of patience, there's not long to go.

 

It will soon be the 10th anniversary of the publication of the first article to describe the Semantic Web. This article signed by Tim Berners-Lee, the inventor of the Web, among others, described a new Web in which the applications were able to understand our wishes, expressed in a natural language, and provide us with the information requested in a fast and efficient way. Since then, several international standards have been developed; true technological pillars on which to base the Web of the future.

The current Web contains billions of documents in different formats. In order to search for the information contained in these documents we have search engines, such as Google, with response times of tenths of a second. But search engines have limitations. One example: if you want to search for “books by García Márquez”, simply entering this text in your favourite search engine will lead you to different documents with the desired information. However, if you wanted to know “what books mention one of García Márquez's books?”, the response would be very different… and erroneous. Why? Because search engines do not understand the semantics of the question. In fact, not even the programmes that index Web documents “understand” the semantics of these documents.

Current searches are based on “key words”. These programmes scrutinise the web in search of new documents, but they do not try to “understand” them, they simple extract their words; although it is true that they are very efficient.

Even though there are already standards to provide this data with semantics, and recently the way of consulting this semantic data has been standardised, it is still very complicated to create web applications that use and generate semantic data.

The work of the researchers at the Universidad Autónoma de Madrid (UAM) and the Polytechnic University (UPM) has created a set of computer tools that make the development of this type of application easier. With these tools, a developer who does not have knowledge of Semantic Web technologies can create web applications that use and generate semantic data. Experimental results show that the web applications generated with this infrastructure have a high level of usability and are satisfactory to users, regardless of their level of computing knowledge.

The benefits of this type of application are not immediate, but nor were those of the first websites. A certain critical mass must be achieved that will compensate the technological investment. Currently, the volume of semantic data on the Internet is much smaller than the volume of documents (non-semantic), but its quality is undoubtedly greater, and this type of application contributes to popularising the use and generation of semantic data. It is worth highlighting that the web applications developed with this infrastructure do not change in appearance or functionality in relation to traditional web applications, so that the user of these applications is not conscious that they are using or generating semantic information.

Although great progress has been made, there is still a lot of work to be done before we can enjoy the world-wide semantic web that the father of the Web outlined. Aspects such as natural language processing or interaction with semantic data are highly complex and have not yet left the research labs. A bit of patience, there's not long to go.

 

It will soon be the 10th anniversary of the publication of the first article to describe the Semantic Web. This article signed by Tim Berners-Lee, the inventor of the Web, among others, described a new Web in which the applications were able to understand our wishes, expressed in a natural language, and provide us with the information requested in a fast and efficient way. Since then, several international standards have been developed; true technological pillars on which to base the Web of the future.

The current Web contains billions of documents in different formats. In order to search for the information contained in these documents we have search engines, such as Google, with response times of tenths of a second. But search engines have limitations. One example: if you want to search for “books by García Márquez”, simply entering this text in your favourite search engine will lead you to different documents with the desired information. However, if you wanted to know “what books mention one of García Márquez's books?”, the response would be very different… and erroneous. Why? Because search engines do not understand the semantics of the question. In fact, not even the programmes that index Web documents “understand” the semantics of these documents.

Current searches are based on “key words”. These programmes scrutinise the web in search of new documents, but they do not try to “understand” them, they simple extract their words; although it is true that they are very efficient.

Even though there are already standards to provide this data with semantics, and recently the way of consulting this semantic data has been standardised, it is still very complicated to create web applications that use and generate semantic data.

The work of the researchers at the Universidad Autónoma de Madrid (UAM) and the Polytechnic University (UPM) has created a set of computer tools that make the development of this type of application easier. With these tools, a developer who does not have knowledge of Semantic Web technologies can create web applications that use and generate semantic data. Experimental results show that the web applications generated with this infrastructure have a high level of usability and are satisfactory to users, regardless of their level of computing knowledge.

The benefits of this type of application are not immediate, but nor were those of the first websites. A certain critical mass must be achieved that will compensate the technological investment. Currently, the volume of semantic data on the Internet is much smaller than the volume of documents (non-semantic), but its quality is undoubtedly greater, and this type of application contributes to popularising the use and generation of semantic data. It is worth highlighting that the web applications developed with this infrastructure do not change in appearance or functionality in relation to traditional web applications, so that the user of these applications is not conscious that they are using or generating semantic information.

Although great progress has been made, there is still a lot of work to be done before we can enjoy the world-wide semantic web that the father of the Web outlined. Aspects such as natural language processing or interaction with semantic data are highly complex and have not yet left the research labs. A bit of patience, there's not long to go.

 

It will soon be the 10th anniversary of the publication of the first article to describe the Semantic Web. This article signed by Tim Berners-Lee, the inventor of the Web, among others, described a new Web in which the applications were able to understand our wishes, expressed in a natural language, and provide us with the information requested in a fast and efficient way. Since then, several international standards have been developed; true technological pillars on which to base the Web of the future.

The current Web contains billions of documents in different formats. In order to search for the information contained in these documents we have search engines, such as Google, with response times of tenths of a second. But search engines have limitations. One example: if you want to search for “books by García Márquez”, simply entering this text in your favourite search engine will lead you to different documents with the desired information. However, if you wanted to know “what books mention one of García Márquez's books?”, the response would be very different… and erroneous. Why? Because search engines do not understand the semantics of the question. In fact, not even the programmes that index Web documents “understand” the semantics of these documents.

Current searches are based on “key words”. These programmes scrutinise the web in search of new documents, but they do not try to “understand” them, they simple extract their words; although it is true that they are very efficient.

Even though there are already standards to provide this data with semantics, and recently the way of consulting this semantic data has been standardised, it is still very complicated to create web applications that use and generate semantic data.

The work of the researchers at the Universidad Autónoma de Madrid (UAM) and the Polytechnic University (UPM) has created a set of computer tools that make the development of this type of application easier. With these tools, a developer who does not have knowledge of Semantic Web technologies can create web applications that use and generate semantic data. Experimental results show that the web applications generated with this infrastructure have a high level of usability and are satisfactory to users, regardless of their level of computing knowledge.

The benefits of this type of application are not immediate, but nor were those of the first websites. A certain critical mass must be achieved that will compensate the technological investment. Currently, the volume of semantic data on the Internet is much smaller than the volume of documents (non-semantic), but its quality is undoubtedly greater, and this type of application contributes to popularising the use and generation of semantic data. It is worth highlighting that the web applications developed with this infrastructure do not change in appearance or functionality in relation to traditional web applications, so that the user of these applications is not conscious that they are using or generating semantic information.

Although great progress has been made, there is still a lot of work to be done before we can enjoy the world-wide semantic web that the father of the Web outlined. Aspects such as natural language processing or interaction with semantic data are highly complex and have not yet left the research labs. A bit of patience, there's not long to go.

 [Contenido incluido(Id:1242651265988;Tipo:UAM_Multimedia_FA)]