Classifying Grains Using Behaviour-Informed Machine Learning

Sorting granular materials such as ores, coffee beans, cereals, gravels and pills is essential forapplications in mineral processing, agriculture and waste recycling. Existing sorting methods are based on the detection of contrast in grain properties including size, colour, density and chemical composition. However, many grain properties cannot be directly detected in-situ, which significantly impairs sorting efficacy. We show here that a simple neural network can infer contrast in a wide range of grain properties by detecting patterns in their observable kinematics. These properties include grain size, density, stiffness, friction, dissipation and adhesion. This method of classification based on behaviour can significantly widen the range of granular materials that can be sorted. It can similarly be applied to enhance the sorting of other particulate materials including cells and droplets in microfluidic devices.

Place cells may simply be memory cells: Memory compression leads to spatial tuning and history dependence

The observation of place cells has suggested that the hippocampus plays a special role in encoding spatial information. However, place cell responses are modulated by several nonspatial variables and reported to be rather unstable. Here, we propose a memory model of the hippocampus that provides an interpretation of place cells consistent with these observations. We hypothesize that the hippocampus is a memory device that takes advantage of the correlations between sensory experiences to generate compressed representations of the episodes that are stored in memory. A simple neural network model that can efficiently compress information naturally produces place cells that are similar to those observed in experiments. It predicts that the activity of these cells is variable and that the fluctuations of the place fields encode information about the recent history of sensory experiences. Place cells may simply be a consequence of a memory compression process implemented in the hippocampus.

Simple phasor-based deep neural network for fluorescence lifetime imaging microscopy

Fluorescence lifetime imaging microscopy (FLIM) is a powerful technique to probe the molecular environment of fluorophores. The analysis of FLIM images is usually performed with time consuming fitting methods. For accelerating this analysis, sophisticated deep learning architectures based on convolutional neural networks have been developed for restrained lifetime ranges but they require long training time. In this work, we present a simple neural network formed only with fully connected layers able to analyze fluorescence lifetime images. It is based on the reduction of high dimensional fluorescence intensity temporal decays into four parameters which are the phasor coordinates, the mean and amplitude-weighted lifetimes. This network called Phasor-Net has been applied for a time domain FLIM system excited with an 80 MHz laser repetition frequency, with negligible jitter and afterpulsing. Due to the restricted time interval of 12.5 ns, the training range of the lifetimes was limited between 0.2 and 3.0 ns; and the total photon number was lower than 106, as encountered in live cell imaging. From simulated biexponential decays, we demonstrate that Phasor-Net is more precise and less biased than standard fitting methods. We demonstrate also that this simple architecture gives almost comparable performance than those obtained from more sophisticated networks but with a faster training process (15 min instead of 30 min). We finally apply successfully our method to determine biexponential decays parameters for FLIM experiments in living cells expressing EGFP linked to mCherry and fused to a plasma membrane protein.

Community Formation as a Byproduct of a Recommendation System: A Simulation Model for Bubble Formation in Social Media

We investigate the problem of the formation of communities of users that selectively exchange messages among them in a simulated environment. This closed community can be seen as the prototype of the bubble effect, i.e., the isolation of individuals from other communities. We develop a computational model of a society, where each individual is represented as a simple neural network (a perceptron), under the influence of a recommendation system that honestly forward messages (posts) to other individuals that in the past appreciated previous messages from the sender, i.e., that showed a certain degree of affinity. This dynamical affinity database determines the interaction network. We start from a set of individuals with random preferences (factors), so that at the beginning, there is no community structure at all. We show that the simple effect of the recommendation system is not sufficient to induce the isolation of communities, even when the database of user–user affinity is based on a small sample of initial messages, subject to small-sampling fluctuations. On the contrary, when the simulated individuals evolve their internal factors accordingly with the received messages, communities can emerge. This emergence is stronger the slower the evolution of individuals, while immediate convergence favors to the breakdown of the system in smaller communities. In any case, the final communities are strongly dependent on the sequence of messages, since one can get different final communities starting from the same initial distribution of users’ factors, changing only the order of users emitting messages. In other words, the main outcome of our investigation is that the bubble formation depends on users’ evolution and is strongly dependent on early interactions.

Rent index forecasting through neural networks

PurposeChinese housing market has been growing fast during the past decade, and price-related forecasting has turned to be an important issue to various market participants, including the people, investors and policy makers. Here, the authors approach this issue by researching neural networks for rent index forecasting from 10 major cities for March 2012 to May 2020. The authors aim at building simple and accurate neural networks to contribute to pure technical forecasting of the Chinese rental housing market.Design/methodology/approachTo facilitate the analysis, the authors examine different model settings over the algorithm, delay, hidden neuron and data spitting ratio.FindingsThe authors reach a rather simple neural network with six delays and two hidden neurons, which leads to stable performance of 1.4% average relative root mean square error across the ten cities for the training, validation and testing phases.Originality/valueThe results might be used on a standalone basis or combined with fundamental forecasting to form perspectives of rent price trends and conduct policy analysis.

Mining hidden knowledge: Embedding models of cause-effect relationships curated from the biomedical literature

We explore the use of literature-curated signed causal gene expression and gene-function relationships to construct un-supervised embeddings of genes, biological functions, and diseases. Our goal is to prioritize and predict activating and inhibiting functional associations of genes, and to discover hidden relationships between functions. As an application, we are particularly interested in the automatic construction of networks that capture relevant biology in a given disease context. We evaluated several unsupervised gene embedding models leveraging literature-curated signed causal gene expression findings. Using linear regression, it is shown that, based on these gene embeddings, gene-function relationships can be predicted with about 95% precision for the highest scoring genes. Func- tion embedding vectors, derived from parameters of the linear regression model, allow to infer relationships between different functions or diseases. We show for several diseases that gene and function embeddings can be used to recover key drivers of pathogenesis, as well as underlying cellular and physiological processes. These results are presented as disease-centric net- works of genes and functions. To illustrate the applicability of the computed gene and function embeddings to other machine learning tasks we expanded the embedding approach to drug molecules, and used a simple neural network to predict drug- disease associations.

Sentiment Analysis of Multilingual Tweets based on Natural Language Processing (NLP)

Multilingual Sentiment analysis plays an important role in a country like India with many languages as the style of expression varies in different languages. The Indian people speak in total 22 different languages and with the help of Google Indic keyboard people can express their sentiments i.e reviews about anything in the social media in their native language from individual smart phones. It has been found that machine learning approach has overcome the limitations of other approaches. In this paper, a detailed study has been carried out based on Natural Language Processing (NLP) using Simple Neural Network (SNN) ,Convolutional Neural Network(CNN), and Long Short Term Memory (LSTM)Neural Network followed by another amalgamated model adding a CNN layer on top of the LSTM without worrying about versatility of multilingualism. Around 4000 samples of reviews in English, Hindi and in Bengali languages are considered to generate outputs for the above models and analyzed. The experimental results on these realistic reviews are found to be effective for further research work.

Sentiment Analysis of Multilingual Tweets Based on Natural Language Processing (NLP)

Multilingual Sentiment analysis plays an important role in a country like India with many languages as the style of expression varies in different languages. The Indian people speak in total 22 different languages and with the help of Google Indic keyboard people can express their sentiments i.e reviews about anything in the social media in their native language from individual smart phones. It has been found that machine learning approach has overcome the limitations of other approaches. In this paper, a detailed study has been carried out based on Natural Language Processing (NLP) using Simple Neural Network (SNN) ,Convolutional Neural Network(CNN), and Long Short Term Memory (LSTM)Neural Network followed by another amalgamated model adding a CNN layer on top of the LSTM without worrying about versatility of multilingualism. Around 4000 samples of reviews in English, Hindi and in Bengali languages are considered to generate outputs for the above models and analyzed. The experimental results on these realistic reviews are found to be effective for further research work.

Using neural networks to support high-quality evidence mapping

Background The Living Evidence Map Project at the Norwegian Institute of Public Health (NIPH) gives an updated overview of research results and publications. As part of NIPH’s mandate to inform evidence-based infection prevention, control and treatment, a large group of experts are continously monitoring, assessing, coding and summarising new COVID-19 publications. Screening tools, coding practice and workflow are incrementally improved, but remain largely manual. Results This paper describes how deep learning methods have been employed to learn classification and coding from the steadily growing NIPH COVID-19 dashboard data, so as to aid manual classification, screening and preprocessing of the rapidly growing influx of new papers on the subject. Our main objective is to make manual screening scalable through semi-automation, while ensuring high-quality Evidence Map content. Conclusions We report early results on classifying publication topic and type from titles and abstracts, showing that even simple neural network architectures and text representations can yield acceptable performance.