Emerging Trends: A new feature from Natural Language Engineering
The Journal of Natural Language Engineering (JNLE) is a true success story. Indeed, the journal has gone from strength to strength in recent years. The impact factor of the journal has increased for the second consecutive year, with the journal being listed among both Linguistics and Computer Science categories. Against the background of a record number of submissions, JNLE has, since 2016, been offering six 160-page issues per year, which by far exceeds the four 96-page issues offered less than 10 years ago!
Speaking about positive recent developments, a distinctive feature I would like to highlight is Robert Dale’s Industry Watch (IW) column. Robert’s fascinating material and writing style have always been attractive to our readers. What appeals to a wide audience is not only his erudition and unique, easy-to-read style, but also the careful selection of very popular topics which bridge the gap between academia and industry. I would advise JNLE readers not to miss any of Robert’s IW pieces and I am sure they have enjoyed his latest one in 23.2: NLP in a Post-Truth World. We are privileged to have Robert Dale writing the IW column which has clearly contributed to the continued success of our journal.
As Executive Editor of the Journal of Natural Language Engineering, I am deeply passionate about making the journal an even greater success.
Over the past two years, I have put a lot of thought into building a strategy to make the journal more successful and popular than ever before. Robert Dale’s column has been highly inspirational and has motivated me to seek further opportunities emulating his successful IW column. I asked myself: why not launch a new column which outlines the latest emerging trends in the field and draws attention to topics and ideas which hold promise for the future? I spoke to Patrick McCartan, Publishing Director at CUP and the Editor then in charge of the journal (who has always been very supportive), and he liked the idea. I came up with the name ‘Emerging Trends’. The next thing I had to do was to think of somebody who could kick-start the column and, even better, take charge of it completely. Several names came to mind, but the two that most stood out were Ed Hovy and Ken Church. Both are big names in Natural Language Processing (NLP), extremely knowledgeable and leading figures in their own areas of expertise. I had a chat with Ed, who had already given me valuable advice on the forthcoming second and substantially revised edition of the Oxford Handbook of Computational Linguistics. Ed made a suggestion, which was perfectly in line with my thinking – Ken Church. I knew that Ken would do an excellent job – not only is he a very well-known expert and big name in statistical NLP, but he is also somebody who is remarkably prescient, with an apparent ability to see far into the future. In addition, his unique, engaging and informal style strikes a very good rapport with the audience.
The Emerging Trends (ET) column has become a reality, as Ken Church kindly agreed to take charge of it. Ken’s ET column discusses emerging trends, which have been gaining traction in recent years and have been identified as bearing high strategic importance within Natural Language Processing, and cutting-edge developments in the field that have the potential to achieve great impact. This column also engages in insightful discussion of recent trends in research practices, which makes for even more exciting reading. The pieces are written in an easy-to-follow and accessible style. In the next few paragraphs, I would like to elaborate on this recently launched column.
In his inaugurating ET piece The Next Generation which was published in issue 22.6, Ken writes that it is easy to suggest where trends come from (the next generation), and how to distinguish passing fads (bubbles) from emerging trends. Young people are often the early adopters, the first to see what is about to happen, but most people do not have the same visionary skills until well after an event has already taken place. Those with the most to lose (the establishment) tend to be the most resistant to change. In this piece Ken goes on to say that many people find Kuhn’s (2012) The structure of scientific revolutions very useful for predicting the future. Kuhn writes that an emerging trend needs to have two qualities. First, it needs to have a few (promising, if not convincing) initial successes that excite a target audience of early adopters (students). Secondly, Kuhn claims that those initial successes should not be too great (which may come across as somehow counter-intuitive) so that there is plenty of room for the next generation to contribute. Ken Church explains that students get to set the direction of a new trend by writing the next round of PhD theses. However, they do not necessarily have the last word on the subject.
In his second ET piece Word2Vec which appeared in 23.1, Ken Church discusses word2vec – an emerging methodology which has recently become very popular. Ken ‘formally proves’ that word2vec is an emerging trend by showing how it satisfies both of Kuhn’s conditions for emerging trends: (i) a few initial (promising, if partial) successes that motivate early adopters (students) to do more, whilst (ii) leaving plenty of room for early adopters to contribute and benefit by doing so. Ken explains that the fact that Google has so much to say on ‘How does word2vec work?’ makes it clear that the definitive answer to that question has yet to be written.
As for the forthcoming ET piece in 23.3: this is something that JNLE readers have to look forward to. Entitled I Did It, I Did It, I Did It, But…, the piece will cover several timeless issues related to research practices. To start with, it will criticise the lack of transparency in some publications which say little about methodology. It will express disquiet at the fact that nowadays many researchers pay lip service to numbers when evaluation results are what matters, and little is said about how things work. Ken argues that there has been a trend for publications to report better and better numbers, but less and less insight. He writes that the literature is turning into a giant leader board, where publication depends on numbers and little else (such as insight and explanation). The forthcoming 23.3 ET feature also offers extensive discussion on further related topics. This is an exciting piece and the fact that readers may not agree with all its statements will make it even more interesting.
Looking further into the future, in 23.5 Ken intends to discuss the ‘inflationary trend with exponential growth’ of papers each year which results from more and more pressure on researchers to publish.
Both the IW column and ET column are real assets to the journal (led by Robert and Ken, who are master columnists)… A regular JNLE issue will feature either an IW or an ET piece. I have already agreed a schedule with both Robert Dale and Ken Church. Although these columns will not appear in special issues, I can safely say that we are all very much looking forward to many excellent IW and ET pieces in the years to come.
JNLE will continue to go from strength to strength. My passion (and ideas) for the journal will not stop here…
Prof Ruslan Mitkov, Executive Editor