Computer translations

1988/06/01 Elhuyar Zientzia Iturria: Elhuyar aldizkaria

• Softwarea

On many occasions we have been able to read articles on this topic, but the main feature of the MicroCATS program that we now analyze is that it is designed for personal computers and not for systems with great advances in artificial intelligence.

Until recently, the only help that the computer could offer when making translations was to provide dictionaries, which is the most widely used or extended today.

Once some dictionaries are inserted in the computer, it is enough to give a word in a language and say to what language we want to translate (these languages should be included in the dictionary) to get the payment of the word entered. However, the response time of these systems is not always as fast as desired and we often turn to printed dictionaries.

New systems, programs capable of translating complete texts, etc. are currently being developed. In the beginning the texts were translated literally and the resulting translations were not very reliable, since when translating a word you have to take into account the context. A step has already been taken and currently translations are made in phrases.

The aforementioned MicroCATS system is one of the following. In addition to the dictionary that is purchased with the program, the user can also create his own dictionaries, such as the one he wants to use to perform a specific job.

When translating a text we can define the search sequence, that is, the preferences between different dictionaries so that you know where to start consulting the system. However, it is logical to think that the first translations will not be very excellent, but that is the development of this system. The translations performed by the computer in its own way should be reviewed and corrected for errors detected. This enriches the knowledge of the computer.

For all this, this system is not yet widespread, but the road is open and you will have to see what you can reach.