Pretrained Transformer models have emerged as state-of-the-art approaches that learn contextual information from text to improve the performance of several NLP tasks. These models, albeit powerful, still require specialized knowledge in specific scenarios. In this paper, we argue that context derived from a knowledge graph (in our case: Wikidata) provides enough signals to inform pretrained transformer models and improve their performance for named entity disambiguation (NED) on Wikidata KG. We further hypothesize that our proposed KG context can be standardized for Wikipedia, and we evaluate the impact of KG context on state-of-the-art NED model for the Wikipedia knowledge base. Our empirical results validate that the proposed KG context can be generalized (for Wikipedia), and providing KG context in transformer architectures considerably outperforms the existing baselines, including the vanilla transformer models.

Advancements in modern technology and computer software have dramatically increased the demand and feasibility to collect high-dimensional data sets. Such data possess challenges which require the creation of new and adaptation of existing statistical methods. One such challenge is that we may observe many more predictors, p, than the number of observations, n, especially in small sample size studies. These data structures are known as high-dimensional, low sample size (HDLSS) data sets, or "small n, big p ". HDLSS data emerge frequently in many health-related fields. For example, in genomic studies, a single microarray experiment might produce tens of thousands of gene expressions compared to the few samples studied, often being less than a hundred (Alag, 2019).

UK-based autonomous vehicle (AV) software company Oxbotica reports it has'mastered' driving on the right-hand side of public roads in Germany after securing AV permit recommendation from the independent inspection body TÜV SÜD. According to Oxbotica, having completed numerous trials on the left-hand side of the road in the UK, including on the complex streets of London, the German AV permit means its AV software has proven itself capable of following the rules of the road and driving on the right in real-world conditions. The official trials started last month on public roads near Friedrichshafen, southern Germany, with a fleet of self-driving vehicles navigating a complex urban environment. To gain the AV permit recommendation, the software company said it had to meet a rigorous assessment framework including detailed hazard analysis and the combination of physical real-world tests and scenario-based simulations. Around two-thirds of the world's population live in countries where cars drive on the right-hand side, including mainland Europe, the US and China.

The Thief Orienteering Problem (ThOP) is a multi-component problem that combines features of two classic combinatorial optimization problems: Orienteering Problem and Knapsack Problem. The ThOP is challenging due to the given time constraint and the interaction between its components. We propose an Ant Colony Optimization algorithm together with a new packing heuristic to deal individually and interactively with problem components. Our approach outperforms existing work on more than 90% of the benchmarking instances, with an average improvement of over 300%.

Relevance Feedback in Content-Based Image Retrieval is a method where the feedback of the performance is being used to improve itself. Prior works use feature re-weighting and classification techniques as the Relevance Feedback methods. This paper shows a novel addition to the prior methods to further improve the retrieval accuracy. In addition to all of these, the paper also shows a novel idea to even improve the 0-th iteration retrieval accuracy from the information of Relevance Feedback.

Researchers from the Queensland University of Technology (QUT) in Australia have developed an algorithm that detects misogynistic content on Twitter. The team developed the system by first mining 1 million tweets. They then refined the dataset by searching the posts for three abusive keywords: whore, slut, and rape. Next, they categorized the remaining 5,000 tweets as either misogynistic or not, based on their context and intent. These labeled tweets were then fed to a machine learning classifier, which used the samples to create its own classification model.

Artificial intelligence technology developed by a Queensland researcher to identify hoax calls is set to be tested at some emergency centres, potentially freeing up operators to deal with real emergencies. University of Southern Queensland computer scientist Dr Rajib Rana has spent the past three years developing the artificial intelligence algorithms required to detect whether someone is genuinely in trouble or whether they are playing a time-wasting prank. Dr Rajib Rana has received an Advance Queensland COVID-19 Industry Research Fellowship to develop his distress inference system. Dr Rana said the "distress inference system" was designed to detect the level of distress in a person's voice, and assess whether it is in line with the sort of incident they are describing. "When people are in real distress there are physiological changes which happen in speech production, like your mouth dries out, your breathing rate increases, that sort of thing," he said.

Refused transactions are frustrating enough for cardholders and merchants, but during a pandemic they could be especially nerve-wracking. Visa Inc. on Wednesday announced a new service for stand-in processing that the payments network says should yield faster and more accurate results when issuers' systems are down. The new Smarter STIP (for Stand-in Processing) technology, which will debut in October, relies on artificial intelligence and so-called deep learning to help reach go or no-go decisions during service interruptions. The new system, which builds on Visa's existing STIP service, has the potential to reduce declines by as much as 50% "in some cases," Visa says. Transaction approvals satisfy cardholders and merchants but also help preserve interchange revenue for issuers.

AI Business Directory is a list of key companies (including startups and big corporations) worldwide with products, services, and applications in the fields related to the Artificial Intelligence (AI). A registered user can submit a listing and maintain it for your own business. The listing service is free. Typical AI fields include, but not limited to: Machine Learning (ML), Deep Learning, Cognitive Computing, Natural Language Processing (NLP), Computer Vision, Pattern Recognition, Autonomous Agents and Multi-Agent Systems, Automated Planning and Scheduling, Robotics, Predictive Analytics, etc. Typical AI applications include, but not limited to: Smart Agriculture, Healthcare, Manufacturing, Smart Cities, Smart Grids, Smart Mobility, Smart Lighting, Smart Buildings, Smart Home, Autonomous Vehicles, Supply Chain and Logistics, Cybersecurity, etc.

The guidance of a large swarm is a challenging control problem. Shepherding offers one approach to guide a large swarm using a few shepherding agents (sheepdogs). Noise is an inherent characteristic in many real-world problems. However, the impact of noise on shepherding is not well-studied. This impact could take two forms. First, noise in the sensorial information received by the shepherd about the location of sheep. Second, noise in the ability of the sheepdog to influence sheep due to disturbances caused during actuation. We study both types of noise in this paper. In this paper, we investigate the performance of Str\"{o}mbom\textquoteright s approach under actuation and perception noises. Before studying the effect noise, we needed to ensure that the parameterisation of the algorithm corresponds to a stable performance for the algorithm. This pegged for running a large number of simulations, while increasing the number of random episodes until stability is achieved. We then systematically studies the impact of sensorial and actuation noise on performance. Str\"{o}mbom\textquoteright s approach is found to be more sensitive to actuation noise than perception noise. This implies that it is more important for the shepherding agent to influence the sheep more accurately by reducing actuation noise than attempting to reduce noise in its sensors. Moreover, different levels of noise required different parameterisation for the shepherding agent, where the threshold needed by an agent to decide whether or not to collect astray sheep is different for different noise levels.