Introduction With The Development Of Big Data And Cloud Comp ✓ Solved

Introduction With the development of big data, cloud computing

With the development of big data, cloud computing, and artificial intelligence, many technological innovations have emerged, and people's lives have become more convenient. Machine translation is one of the most important technologies. It refers to the automation technology that can translate oral or written text from one language to another without human participation. As the Internet has opened up a wider multilingual world for people, this language service has become very valuable. In the past few years, machine translation research and development has been remarkable. Back in 2016, Google Translate launched neural machine translation, which uses phrase-based machine translation to reduce the gap between human translation and machine translation.

This review will focus on two questions based on the development of Google Translate: 1) How does Google Translate work? 2) Will it replace human translation?

Machine Translation: Google Translate

Machine translation is a method that relies on computer technology and information technology. The principles of translation play an important role. Methods of machine translation include literal translation, the conversion method achieved through cohesion and conversion, and the intermediate language method that achieves the purpose of translation through feedback from the intermediate language. The principle of these methods is to analyze and study the languages to achieve the translation's objective. Different methods have varying levels of interpretation.

Literal translation is a direct translation method that does not require extensive language analysis, making it the most common method in machine translation. It is a method based on the comparison level in translation work. The latter two methods require a certain degree of language analysis to obtain translation results.

Neural Machine Translation (NMT)

According to a Sure-Language website, Google Translate primarily utilizes Neural Machine Translation (NMT). NMT employs neural network technology to achieve contextually accurate translation, rather than translating broken sentences word by word. Using a large artificial neural network, NMT calculates the probability of word sequences and processes entire sentences as a cohesive model. NMT learns and collects information to imitate human brain neurons, establishing connections and evaluating input as a whole unit. The process is divided into encoding and decoding stages. In the encoding stage, the source language text is inputted and categorized into language vectors, placing similar-context words into comparable word vectors. The decoding stage then effectively sends the vector to the target language. Throughout the process, technology transcends simple word translation; it also translates context and information.

However, NMT cannot guarantee 100% accuracy. A case study comparing NMT and Statistical Machine Translation (SMT) shows that NMT has a higher chance of producing mistranslation errors (Pinnis & Skadiņa, 2017), affecting machine translation's overall accuracy.

Challenges of Machine Translation

Machine translation is fast, convenient, and cost-effective, capable of translating vast amounts of text in seconds. It is free for non-informal translations and offers hundreds of applications that can translate text, images, and even voice at the press of a finger. Additionally, machine translators can handle hundreds of languages, achieving feats beyond human capability.

However, significant weaknesses remain in machine translation. Firstly, machine translation lacks cultural sensitivity, as it cannot recognize slang, jargon, puns, and idioms. For instance, the phrase “人来人往” (ren lai ren wang) means "prosperous" in Chinese, and while Google Translate captures the literal meaning, it fails to interpret the intention behind "People come and go." Different cultures have distinct language systems, making it challenging to program machines to understand or experience specific cultures. Consequently, translated content may not align with cultural values and norms, presenting a significant challenge for machines.

Another limitation is that machine translation struggles to connect words in context. In many languages, the same word can have multiple unrelated meanings, emphasizing the importance of context for understanding. Currently, only humans possess the ability to combine words with their contextual meanings effectively, which gives them an advantage in translation tasks.

Conclusion

Machine translation can replace humans in certain tasks that require speed and massive output. However, it is unlikely that it will achieve the quality required in professional translation scenarios, particularly in specialized fields. Full automation in translation will likely remain confined to narrow domains for the foreseeable future. Ultimately, the effectiveness of human translation surpasses that of machine translation, particularly regarding accuracy. Thus, while machine translation has its advantages, combining machine and human translations can create a win-win situation. This synergy not only saves time and ensures quality but aligns with trends in future development.

References

• Pinnis, M., & Skadiņa, I. (2017). NMT or SMT: Case Study of a Narrow-domain English-Latvian Post-editing Project.
• Costa-jussa, Marta & Rapp, Reinhard & Lambert, Patrik & Eberle, Kurt & Banchs, Rafael & Babych, Bogdan. (2016). Hybrid Approaches to Machine Translation.
• Hao, K. (2020). AI still doesn't have the common sense to understand human language.
• Hintikka, J. (1977). Quantifiers in Natural Languages: Some Logical Problems.
• Golumbic, M. C. (1990). Advances in Artificial Intelligence.
• Paris, C. L. (1991). Natural Language Generation in Artificial Intelligence and Computational Linguistics.
• Liu, B. (2015). Chinese Lexical Semantics. 16th Workshop, CLSW 2015.
• Manchester, W. (2020). The Future of Language Translation: A Perspective.
• Swartout, W. R. (1991). Knowledge Representation in Natural Language Processing.
• Cellan-Jones, R. (2014). Stephen Hawking warns artificial intelligence could end mankind.