Future of Natural Language Processing – Potential Lists of Topics for PhD students

• Stanford Core NLP
• GATE
• NLTK
• Apache OpenNLP
• WEKA
• QTag POS Tagger
•  Morphy
•  Mate Tools
•  MALLET Toolkit
•  Link Grammar Parser
• LOLITA NLP system
• Google Speech API
• Flex/Lex
• Chamiak Parser
• POS Tagger
• Gensim
• Google data API
• IBM Watson System

British

National Corpus

GermanNet

FrameNET

CM-1

DBPedia

Homby’s Verb Patterns

Google News Corpus

MUC Shared Tasks Datasets

Open American National Corpus

Brown Corpus

Stanford CoreNLP

WordNet

LSTM

CNN

RNN

As AI tries to take advantage of the technology’s prospects, NLP would get even more advanced.

1. The massive shift from data-driven to intelligence-driven decision making

Smart officialdoms now make decisions based not on data only but on the intelligence derived from that data by NLP-powered machines.

2. Creation of more extensive, better NLP platforms like Spark NLP

Data scientists dealing with NLP and other AI aspects rely on NLP library platforms to construct and trial their applications. The platform pool such as OpenNMT, Stanford’s CoreNLP, SpaCy, and Tensor Flow has been widely used.

3. Eradication of human data scientists

Data scientists would be wiped out in the future as NLP advances along with Machine Learning, and its features such as pattern recognition, advanced analysis, and interpretation improve beyond today’s level.

4. A swap to natural language

The future test in NPL would be able to understand the human language. In the future, natural language processing would have to evolve in its function to become natural language understanding.

Recent trends in the NLP from Scholarly Papers published in Scopus Indexed Journals

• Recently, analysing the user reviews, Aldabbas et al., (2021) scrapped google play content and knowledge engineering¹.
• Food recipes were altered and generated with NLP techniques by Pan et al., (2020)
• Construct identity problem were tackled using NLP techniques by Ludwig et al., (2020)².
• Multi-class Categorization of design build contract technique requirement were built using text mining and NLP by Hassan et al., (2020)³
• Building personalized educational material for chronic disease patients by Wang et al., (2020).
• Medical based NLP techniques: Clinical decision making with EHRs and intelligent patient summaries using ML and NLP techniques by Trappey et al., (2020). In medical, the current deep learning-based NLP techniques focus into three major purposes: representation learning, information extraction and clinical prediction.
• Analysing news articles using NLP by Titiliuc, Ruseti, & Dascalu (2020) 4– Semantic similarities between articles and rank various publications based on their influences – Visualization to ease understanding. Techniques such as opinion mining, geographical name extraction nd content quality assessment are the future scope.

Data Sets for NLP

1. Feedback from users in app stores
2. Social media and developer comments in discussion forums
3. Patient story – qualitative interview, voice recording
4. Newspaper articles
5. Customer review
6. Social media comments
7. Stories
8. Content from the Manuscripts / Journal articles and many more

Conclusion

NLP can analyse and bond with language-based information by making machines equipped to understand the content and substitute human tasks like abstracting, translation, classification, and mining. Moreover, NLP giving organizations a way to analyze shapeless information, customer supports communications, product analyses, and social media messages. So there are many Research Gap is yet to be determined in this field. Hence it would be a good opening for the researchers to start research in this area.

References

1. Aldabbas, H., Bajahzar, A., Alruily, M., Qureshi, A. A., Latif, R. M. A., & Farhan, M. (2021). Google Play Content Scraping and Knowledge Engineering using Natural Language Processing Techniques with the Analysis of User Reviews. J Intell Syst, 30(1), 192–208.
2. Y. Pan, Q. Xu and Y. Li, “Food Recipe Alternation and Generation with Natural Language Processing Techniques,” 2020 IEEE 36th International Conference on Data Engineering Workshops (ICDEW), Dallas, TX, USA, 2020, pp. 94-97, doi: 10.1109/ICDEW49219.2020.000-1.
3.  Fahad ul Hassan; Tuyen Le; and Duc-Hoc Tran, Multi-Class Categorization of Design-Build Contract Requirements Using Text Mining and Natural Language Processing Techniques, Construction Research Congress 2020,
4. C. Titiliuc, S. Ruseti, and M. Dascalu, “What’s Been Happening in the Romanian News Landscape? A Detailed Analysis Grounded in Natural Language Processing Techniques,” 2020 22nd International Symposium on Symbolic and Numeric Algorithms for Scientific Computing (SYNASC), Timisoara, Romania, 2020, pp. 195-201, doi: 10.1109/SYNASC51798.2020.00040.