Evaluating disfluency removal in speech requires more than aggregate token-level scores. Traditional word-based metrics such as precision, recall, and F1 (E-Scores) capture overall performance but cannot reveal why models succeed or fail. We introduce Z-Scores, a span-level linguistically-grounded evaluation metric that categorizes system behavior across distinct disfluency types (EDITED, INTJ, PRN). Our deterministic alignment module enables robust mapping between generated text and disfluent transcripts, allowing Z-Scores to expose systematic weaknesses that word-level metrics obscure. By providing category-specific diagnostics, Z-Scores enable researchers to identify model failure modes and design targeted interventions -- such as tailored prompts or data augmentation -- yielding measurable performance improvements. A case study with LLMs shows that Z-Scores uncover challenges with INTJ and PRN disfluencies hidden in aggregate F1, directly informing model refinement strategies.

EDITED is the global leader in Retail Intelligence. We help retailers increase margins, generate more sales and drive better outcomes through AI-driven Automation and Market & Enterprise Intelligence. By connecting internal enterprise and external market data, brands like John Lewis and Puma use the EDITED suite of intelligence products to drive insights into action in a more profitable and coordinated way. At EDITED, we always dream big, with a commitment to deliver measurable value to our customers. We believe in exceeding expectations and challenging the status quo. We do so by delivering our powerful Retail Intelligence Platform that drives better and faster decision making for retailers.

One of the grand challenges discussed during the Dagstuhl Seminar "Knowledge Graphs: New Directions for Knowledge Representation on the Semantic Web" and described in its report is that of a: "Public FAIR Knowledge Graph of Everything: We increasingly see the creation of knowledge graphs that capture information about the entirety of a class of entities. [...] This grand challenge extends this further by asking if we can create a knowledge graph of "everything" ranging from common sense concepts to location based entities. This knowledge graph should be "open to the public" in a FAIR manner democratizing this mass amount of knowledge." Although linked open data (LOD) is one knowledge graph, it is the closest realisation (and probably the only one) to a public FAIR Knowledge Graph (KG) of everything. Surely, LOD provides a unique testbed for experimenting and evaluating research hypotheses on open and FAIR KG. One of the most neglected FAIR issues about KGs is their ongoing evolution and long term preservation. We want to investigate this problem, that is to understand what preserving and supporting the evolution of KGs means and how these problems can be addressed. Clearly, the problem can be approached from different perspectives and may require the development of different approaches, including new theories, ontologies, metrics, strategies, procedures, etc. This document reports a collaborative effort performed by 9 teams of students, each guided by a senior researcher as their mentor, attending the International Semantic Web Research School (ISWS 2019). Each team provides a different perspective to the problem of knowledge graph evolution substantiated by a set of research questions as the main subject of their investigation. In addition, they provide their working definition for KG preservation and evolution.

"Taken together, [studies show] internet addiction is associated with structural and functional changes in brain regions involving emotional processing, executive attention, decision making, and cognitive control." But what about kids who aren't "addicted" per se? Addiction aside, a much broader concern that begs awareness is the risk that screen time is creating subtle damage even in children with "regular" exposure, considering that the average child clocks in more than seven hours a day (Rideout 2010). As a practitioner, I observe that many of the children I see suffer from sensory overload, lack of restorative sleep, and a hyperaroused nervous system, regardless of diagnosis--what I call electronic screen syndrome. These children are impulsive, moody, and can't pay attention--much like the description in the quote above describing damage seen in scans.

AI is transforming how we do business at an unprecedented pace, but the transition to becoming AI-driven is easier than you think. Now is the time to invest and remain at the top of your game. A few weeks ago Artificial Intelligence was thrown into the spotlight as the winners of this year's Turing award, widely known as computing's "Nobel Prize", were announced. Yoshua Bengio, Geoffrey Hinton and Yann LeCun were recognized for their pioneering work over a 30-year period on AI. In the 80s, when AI research was very unfashionable, the trio were motivated by the idea of creating algorithms that could operate, at least at the level of metaphor, like a brain. Neurons in the brain work as an ensemble, together learning internal representations of the input they observe, and these researchers believed their artificial neural networks could do the same.

When Detroit-based luxury goods brand Shinola began working on its new Vinton watch, the team designed with a woman in mind, but testing the product through analytics platform MakerSights, which correlates consumer feedback with historical sales data, revealed the style appealed to all genders. As a result, the brand deepened its buy-in on those by about 70 per cent. "You never design by data," says Shinola CEO Tom Lewand, "but the data provides a compass as you're navigating a hunch." In other words, Shinola already had a great vision – and the data enhanced it. MakerSights is among a new class of data-driven analytics platforms that combine factors such as search queries, social media activity, e-commerce sell-throughs and consumer feedback to provide clues into what is most likely to become a trend.

This piece accompanies a dedicated series from Ben around intelligence, A.I, and data-driven design and development in retail – all of which you can find in our 7th Edition. Similarly, you will find references to other'features', which denote to the other editorial pieces in our 7th Edition Report.] Just as WhichPLM did for both of our previous special editorial examinations (covering 3D in 2015, and the Internet of Things in 2016) the last exclusive feature in our 7th Edition acts as the final piece of the puzzle, collecting guidance, food for thought, and practical recommendations for retailers and brands who may be looking to lay the long-term groundwork for their own A.I. initiatives, or to embark on a particular, more pressing project. The clearest question for prospective customers of A.I. solutions: are these viable products, with clear return on investment potential? Broadly speaking, the answer is yes. While general intelligence – a single machine to run everything, with mental capacities far in excess of our own, across essentially all of human endeavour – remains a pipe dream, more focused applications of narrow, specialised A.I. are limited only by customers' ability to find the right technology partner and to gain access to their own information and broader market data in sufficient volume to deliver results. But even if A.I. was more limited – its capabilities confined to being a better analytics platform or Business Intelligence tool, for instance – I believe it would still rank as an essential investment for many retailers and brands.

Fashion retail is no different; at EDITED we're interested in what these technologies can teach us about our industry and the unique data it produces. While we've got retail data pretty covered, there's a huge amount of imagery in fashion to play with. So we created a specific kind of neural network -- called a convolutional variational autoencoder -- to investigate a dataset of runway photos from London Fashion Week (LFW). The network synthesises information from the photos and then represents them as a set of numbers that can be manipulated. That allowed us to analyse the collection in super interesting ways.

High street retail hasn't changed much in the last few decades. Yep, there's click and collect and online returns but, as in years gone by, product buyers decide what will sell by using a mix of nous and trends analysis. Fashion, for example, may be getting faster (quicker production time and fulfilment) but the knack is still in predicting the season's trends and riding the wave. In-store merchandising, too, is a matter of long-honed instincts as to what should go where. What I'm saying is there's a lot of art in the high-street retail business (particularly fashion), and it attracts suitably artistic people.

If there's any doubt as to why "big data" has become as ubiquitous in business as pens, chairs and coffee mugs, look no further than the margins. Since becoming the buzzword of the decade, big data has given countless businesses huge competitive advantages by redefining the quality of the information at their fingertips and the speed at which they can react. So why have apparel brands lagged in doing the same? In many cases, identifying the "next big thing" -- what will sell, and the rate at which it will fly off the shelves -- is still steeped in guesswork and unsupported instinct. For brands, the use case is abundantly clear.