In the last decade, social networks became most popular medium for communication and interaction. As an example, micro-blogging service Twitter has more than 200 million registered users who exchange more than 65 million posts per day. Users express their thoughts, ideas, and even their intentions through these tweets. Most of the tweets are written informally and often in slang language, that contains misspelt and abbreviated words. This paper investigates the problem of selecting features that affect extracting user's intention from Twitter feeds based on text mining techniques. It starts by presenting the method we used to construct our own dataset from extracted Twitter feeds. Following that, we present two techniques of feature selection followed by classification. In the first technique, we use Information Gain as a one-phase feature selection, followed by supervised classification algorithms. In the second technique, we use a hybrid approach based on forward feature selection algorithm in which two feature selection techniques employed followed by classification algorithms. We examine these two techniques with four classification algorithms. We evaluate them using our own dataset, and we critically review the results.

Applying Q-learning to high-dimensional or continuous action spaces can be difficult due to the required maximization over the set of possible actions. Motivated by techniques from amortized inference, we replace the expensive maximization over all actions with a maximization over a small subset of possible actions sampled from a learned proposal distribution. The resulting approach, which we dub Amortized Q-learning (AQL), is able to handle discrete, continuous, or hybrid action spaces while maintaining the benefits of Q-learning. Our experiments on continuous control tasks with up to 21 dimensional actions show that AQL outperforms D3PG (Barth-Maron et al, 2018) and QT-Opt (Kalashnikov et al, 2018). Experiments on structured discrete action spaces demonstrate that AQL can efficiently learn good policies in spaces with thousands of discrete actions.

FirstWord MedTech's Digital Ten is a fortnightly round-up of the 10 most read and noteworthy headlines related to digital health, including industry deals, alliances, collaborations, innovations and R&D news. The biggest M&A deal inked this year so far comes from the telemedicine field with Teladoc Health agreeing to fork out $600 million to acquire InTouch Health, a provider of cloud-based telemedicine software and physician services for hospitals and health systems. Teladoc's existing telehealth service platform targets consumers and with this deal, it will gain a complementary business that is expected to generate revenue of $80 million in 2019, representing 35% year-over year growth, and a new facility-based virtual care platform. Mojo Vision's smart contact lens receives FDA breakthrough device designation In yet another first for 2020, Mojo Vision's Mojo smart contact lens is the first ophthalmic product to get FDA breakthrough device designation this year. The lens incorporates what the company describes as the "smallest and densest dynamic display ever made," along with a power-efficient image sensor optimised for computer vision, a custom wireless radio, and motion sensors for eye-tracking and image stabilisation.

Diabetes is a global public health disease projected to affect 642 million adults by 2040, with about 75% residing in low- and middle-income countries. Diabetic retinopathy (DR) affects 1 in 3 people with diabetes and remains the leading cause of blindness in working-aged adults. There are 3 broad strategic imperatives to prevent blindness caused by DR. Primary prevention requires preventing or delaying the onset of DR in those with diabetes by systems-level lifestyle modifications such as increasing physical activity or dietary modifications, pharmacological interventions for glycaemic and blood pressure control, and systematic screening for the onset of DR. Secondary prevention requires preventing the progression of DR in patients with DR by continuing systemic risk factor control, regular screening to monitor for the progression of mild DR to vision-threatening stages, and the development and implementation of evidence-based guidelines for managing DR. In this aspect, telemedicine-based DR screening incorporating artificial intelligence technology has the potential to facilitate more widespread and cost-effective screening, particularly in low- and middle-income countries. Tertiary prevention of DR blindness has been the main focus of the clinical ophthalmology community, classically based on laser photocoagulation treatment and ocular surgery but with an increasing use of anti-vascular endothelial growth factor (anti-VEGF) for vision-threatening DR. Evidence from serial epidemiological studies shows blindness due to DR has declined in high-income countries (e.g., the USA and UK) due to coordinated public health education efforts, increased awareness, early detection by DR screening, sustained systemic risk factor control, and the availability of effective tertiary level treatment. However, the progress made in reducing DR blindness in high-income countries may be overwhelmed by the increasing numbers of patients with diabetes and DR in low- and middle-income countries (e.g., China, India, Indonesia, etc.).

Cows chat to each other about food and the weather, according to an astonishing new study that revealed they all have their own unique'moo', writes Laura Sharman. The study revealed how dairy cows respond to positive and negative emotional situations each with their own individual voice and linked their moods to their moos. Experts reckon they can decipher the language of the cows, made up from different-pitched moos, using computer software they dubbed'Google translate for cows.' Biologists made the discovery by listening to Holstein-Fresian heifer cattle, a European breed, mooing into a microphone and analysing the pitch. They found each cow retains its own distinct moo and can give cues in different situations which helps them to maintain contact with the herd and express excitement, arousal, engagement or distress. Lead study author Alexandra Green, from the University of Sydney in Australia, said: "We found that cattle vocal individuality is relatively stable across different emotionally loaded farming contexts. "Recognising individual cattle could assist farmers in the non-invasive detection of welfare." The findings could help farmers keep their cattle healthy and happy by understanding each cow's mood by translating their individual moos. Scientists hope it will also help improve animal welfare thanks to an area of study which historically had not received much attention. Cattle mothers and offspring are known to communicate by maintaining individuality in their lowing, according to existing research. But the new study confirms that cows maintain this individual mooing throughout their lives, even when they are nattering among themselves. Dairy cows communicate with each other all the time, but when they are talking about happier things, like food, their moos are more sonorous. And when they are moaning about the weather, their moos, while still retaining their fingerprint-like individuality, are pitched lower. PhD candidate Alexandra said: "Cows are gregarious, social animals.

I used this example in my talk at useR!2019 in Toulouse, and it is also the basis of a vignette in the package, and a recent blog post by Mitchell O'Hara-Wild. The data set contains domestic tourist visitor nights in Australia, disaggregated by State, Region and Purpose. An example of a feature would be the autocorrelation function at lag 1 -- it is a numerical summary capturing some aspect of the time series. Autocorrelations at other lags are also features, as are the autocorrelations of the first differenced series, or the seasonally differenced series, etc. Another example of a feature is the strength of seasonality of a time series, as measured by \(1-\text{Var}(R_t)/\text{Var}(S_t R_t)\) where \(S_t\) is the seasonal component and \(R_t\) is the remainder component in an STL decomposition.

Emergency Response Management (ERM) is a critical problem faced by communities across the globe. Despite its importance, it is common for ERM systems to follow myopic and straight-forward decision policies in the real world. Principled approaches to aid decision-making under uncertainty have been explored in this context but have failed to be accepted into real systems. We identify a key issue impeding their adoption - algorithmic approaches to emergency response focus on reactive, post-incident dispatching actions, i.e. optimally dispatching a responder after incidents occur. However, the critical nature of emergency response dictates that when an incident occurs, first responders always dispatch the closest available responder to the incident. We argue that the crucial period of planning for ERM systems is not post-incident, but between incidents. However, this is not a trivial planning problem - a major challenge with dynamically balancing the spatial distribution of responders is the complexity of the problem. An orthogonal problem in ERM systems is to plan under limited communication, which is particularly important in disaster scenarios that affect communication networks. We address both the problems by proposing two partially decentralized multi-agent planning algorithms that utilize heuristics and the structure of the dispatch problem. We evaluate our proposed approach using real-world data, and find that in several contexts, dynamic re-balancing the spatial distribution of emergency responders reduces both the average response time as well as its variance.

As the prevalence of deep learning in computer vision, adversarial samples that weaken the neural networks emerge in large numbers, revealing their deep-rooted defects. Most adversarial attacks calculate an imperceptible perturbation in image space to fool the DNNs. In this strategy, the perturbation looks like noise and thus could be mitigated. Attacks in feature space produce semantic perturbation, but they could only deal with low resolution samples. The reason lies in the great number of coupled features to express a high-resolution image. In this paper, we propose Attack by Identifying Effective Features (AIEF), which learns different weights for features to attack. Effective features, those with great weights, influence the victim model much but distort the image little, and thus are more effective for attack. By attacking mostly on them, AIEF produces high resolution adversarial samples with acceptable distortions. We demonstrate the effectiveness of AIEF by attacking on different tasks with different generative models.

Increasingly complex and autonomous systems require machine ethics to maximize the benefits and minimize the risks to society arising from the new technology. It is challenging to decide which type of ethical theory to employ and how to implement it effectively. This survey provides a threefold contribution. Firstly, it introduces a taxonomy to analyze the field of machine ethics from an ethical, implementational, and technical perspective. Secondly, an exhaustive selection and description of relevant works is presented. Thirdly, applying the new taxonomy to the selected works, dominant research patterns and lessons for the field are identified, and future directions for research are suggested.

In such systems the dialogue state tracker (DST) is a core component, aimed to maintain a distribution over the dialogue states based on the dialogue history. A dialogue state at any turn t in the dialogue is typically represented as a set of slot-value pairs, such as ( price, moderate) or ( food, italian) in the context of restaurant reservation. The goal of the DST is to determine the user's intent and the user's goal during the dialogue and represent them as such slot-value pairs. The downstream components of a dialogue system (e.g the dialogue manager) that are responsible to choose the next system action, rely on an accurate DST for an effective dialogue strategy. Because of the importance of DST in dialogue systems, their development attracted lots of research both in academia and industry. Typical dialogue systems are modeled for a fixed ontology consisting of a single domain (Mrk ˇ si c et al. 2017; Zhong, Xiong, and Socher 2018; Ren et al. 2018), and the domain ontology schema defines intents, slots and values for each slot of the domain.