Greece

Natural gas constitutes one of the most actively traded en-
ergy commodity with a significant impact on many financial activities
of the world. The accurate natural gas price prediction and the direc-
tion of price changes are considered essential since these forecasts are
utilized in energy sustainability planning, commodity trading and deci-
sion making, covering both the supply and demand side of natural gas
market. In this research, a new deep learning prediction model is pro-
posed for short-term forecasting natural gas price and movement. The
proposed forecasting model exploits the ability of convolutional layers
for providing a deep insight in natural gas data and the eficiency of
LSTM layers for learning short-term and long-term dependencies. Ad-
ditionally, a significant advantage of the proposed model is its abilities
to predict the price of natural gas on the following day (regression) and
also to predict if the price on the next day will increase, decrease or stay
stable (classfication) with respect to today&#39;s price. The conducted se-
ries of experiments demonstrated that the proposed model considerably
outperforms state-of-the-art deep learning and machine learning models.

AIAI 2020

An advanced deep learning model for short-term forecasting U.S. natural gas price and movement

AIAI 2020 (16th International Conference on Artificial Intelligence Applications and Innovations) should be held in Halkidiki, Greece. Due to current Covid-19 situation the conference will be held virtually. Underline will host a repository of video talks and preserve them for the future. Videos will be released on Underline one day after the conference.

Natural gas constitutes one of the most actively traded en-
ergy commodity with a significant impact on many financial activities
of the world. The accurate natural gas price prediction and the direc-
tion of price changes are considered essential since these forecasts are
utilized in energy sustainability planning, commodity trading and deci-
sion making, covering both the supply and demand side of natural gas
market. In this research, a new deep learning prediction model is pro-
posed for short-term forecasting natural gas price and movement. The
proposed forecasting model exploits the ability of convolutional layers
for providing a deep insight in natural gas data and the eficiency of
LSTM layers for learning short-term and long-term dependencies. Ad-
ditionally, a significant advantage of the proposed model is its abilities
to predict the price of natural gas on the following day (regression) and
also to predict if the price on the next day will increase, decrease or stay
stable (classfication) with respect to today's price. The conducted se-
ries of experiments demonstrated that the proposed model considerably
outperforms state-of-the-art deep learning and machine learning models.

technical paper

As fingerprint recognition systems have become increasingly adopted within a range of technology applications over the last decade, so too has their attention within emerging research. However, although this increased attention has led to an enhancement of the software and algorithms behind this recognition pro-cess, the majority of research has still not addressed the issues of incorrect rotation or proximity between the finger and the device. Current systems assume that the direction of the imprinted finger will align with that of the target fingerprint image; this decreases the accuracy of fingerprint recognition across a variety of finger orientations and scenarios. In response to this use-case dilemma, this paper proposes a new technique of pattern matching that can account for this natural range of fingerprint orientations. This is achieved first through a preliminary stage of orientation identification, whereby the fingerprint image can be stored under multiple permutations by using approximate circular string-matching algorithms. 
This enables the database of images for each approximate permutation of orientation to be stored in advance. It can then be matched against the strong information of the fingerprint at its exact relative rotation of input. The improved accuracy of recognition demonstrated through the results of this study may enable the functionality of fingerprint recognition to adapt to challenging device form-factors and provide the accuracy needed for military and medical applications. 


Fingerprints Recognition System-Based on Mobile Device Identification using Circular String Pattern Matching Techniques

Since the discovery of motifs in molecular sequences for real genomic data, research into this phenomenon has attracted increased attention. Motifs are relatively short sequences that are biologically significant. This paper utilises the bioinformatics application of the (MMDSW) algorithm , testing it using real genomic data from large sequences. It intends to implement bioinformatics application for real genomic data, in order to discover interesting regions for all maximal motifs, in a sliding window of length ℓ, on a sequence x of length n.

Application and Algorithm: Maximal Motif Discovery for Biological Data in a Sliding Window

EU Draft Ethics guidelines for Trustworthy AI has been proposed to promote ethical, lawful and robust AI solutions. In this article, we entertain the systemic issues and challenges of any development of the proposed guidelines.

Trustworthy AI Needs Unbiased Dictators!

The performance of solving a constraint problem can often be improved by converting a subproblem into a single constraint (for example into a regular membership constraint or a table constraint). In the past, it stood out, that specialist constraint solvers (like simplex solver or SAT solver) outperform general constraint solvers, for the problems they can handle. The disadvantage of such specialist constraint solvers is that they can handle only a small subset of problems with special limitations to the domains of the variables and/or to the allowed constraints. In this presentation we introduce the concept of fd-complete constraints and fd-complete constraint satisfaction problems, which allow combining both previous approaches. More accurately, we convert general constraint problems into problems which use only one, respectively one kind of constraint. The goal is it to interpret and solve the converted constraint problems with specialist solvers, which can solve the transformed constraint problems faster than the original solver the original constraint problems.

An Introduction of FD-Complete Constraints

Advancements in Long Short Term Memory (LSTM) Net- works have shown remarkable success in various Natural Language Gen- eration (NLG) tasks. However, generating sequence from pre-specified lexical constraints is a new, challenging and less researched area in NLG. Lexical constraints take the form of words in the language model’s out- put to create fluent and meaningful sequences. Furthermore, most of the previous approaches cater this problem by allowing the inclusion of pre- specified lexical constraints during the decoding process, which increases the decoding complexity exponentially or linearly with the number of constraints. Moreover, some of the previous approaches can only deal with single constraint. Additionally, most of the previous approaches only deal with single constraints. In this paper, we propose a novel neu- ral probabilistic architecture based on backward-forward language model and word embedding substitution method that can cater multiple lexical constraints for generating quality sequences. Experiments shows that our proposed architecture outperforms previous methods in terms of intrinsic evaluation.

Backward-Forward Sequence Generative Network for Multiple Lexical Constraints

As the requirements for automatic speech recognition are continually increasing, the demand for accuracy and efficiency is also of particular interest. In this paper, we present most of the well-known Automated Speech Recognition systems(ASR), and we benchmark three of them, namely the IBM Watson, Google, and Wit, using the WER, Hper, and Rper error metrics. The experimental results show that Google's automatic speech recognition performs better among the three systems. We intend to extend the benchmarking both to include most of the available Automated Speech Recognition systems and increase our test data.

Α Benchmarking of IBM, Google and Wit Automatic Speech Recognition Systems

Traffic load classification for Smart Cities based on weather data collected by sensors. Comparison and evaluation of predictability for periods throughout year.

Using Classification for Traffic Prediction in Smart Cities

Due to the surge of interest in online retailing, the use of credit cards has been rapidly expanded in recent years. Stealing the card details to perform online transactions, which is called fraud, has also seen more frequently. Preventive solutions and instant fraud detection methods are widely studied due to critical financial losses in many industries. In this work, a Gradient Boosting Tree (GBT) model for the real-time detection of credit card frauds on the streaming Card-Not-Present (CNP) transactions is investigated with the use of different attributes of card transactions. Numerical, hand-crafted numerical, categorical and textual attributes are combined to form a feature vector to be used as a training instance. One of the contributions of this work is to employ transaction aggregation for the categorical values and inclusion of vectors from a character level word embedding model which is trained on the merchant names of the transactions. The other contribution is introducing a new strategy for training dataset generation employing the sliding window approach in a given time frame to adapt to the changes on the trends of fraudulent transactions. In the experiments, the feature engineering strategy and the automated training set generation methodology are evaluated on the real credit card transactions.

An Adaptive Approach on Credit Card Fraud Detection using Transaction Aggregation and Word Embeddings

With the advent of social media, concepts such as forecasting and nowcast-ing became part of the public debate. Past successes include predicting elec-tion results, stock prices and forecasting events or behaviors. This work aims at using Twitter data, related to songs and artists that appeared on the top 10 of the Billboard Hot 100 charts, and perform sentiment analysis on the col-lected tweets, to predict the charts in the future. Our goal is to investigate the relation between the number of mentions of a song and its artist, as well as the semantic orientation of the relevant posts and its performance on the subsequent chart. The problem was approached via regression analysis, which estimated the difference between the actual and predicted positions and moderate results. We also focused on forecasting chart ranges, namely the top 5, 10 and 20. Given the accuracy and F-score achieved and com-pared to previous research, our findings are deemed satisfactory, especially for predicting the top 20.

Using Twitter to Predict Chart Position for Songs

The emerging 5G networks promises more throughput, faster, and more reliable services, but as the network complexity and dynamics increases, it becomes more difficult to troubleshoot the systems. Vendors are spending a lot of time and effort on early anomaly detection in their development cycle and majority of the time is spent on manually analyzing system logs. While main research in anomaly detection uses performance metrics, anomaly detection using functional behaviour is still lacking in depth analysis. In this paper we show how a boosted ensemble of Long Short Term Memory classifiers can detect anomalies in the 5G Radio Access Network system logs. Acquiring system logs from a live 5G network is difficult due to confidentiality issues, live network disturbance, and problems to repeat scenarios. Therefore, we perform our evaluation on logs from a 5G test bed that simulate realistic traffic in a city. Our ensemble learns the functional behaviour of an application by training on logs from normal execution time. It can then detect deviations from normal behaviour and also be retrained on false positive cases found during validation. Anomaly detection in RAN shows that our ensemble called BoostLog, outperforms a single LSTM classifier and further testing on HDFS logs confirms that BoostLog also can be used in other domains. Instead of using domain experts to manually analyse system logs, BoostLog can be used by less experienced trouble shooters to automatically detect anomalies faster and more reliable.

An advanced deep learning model for short-term forecasting U.S. natural gas price and movement

Downloads

Next from AIAI 2020

Fingerprints Recognition System-Based on Mobile Device Identification using Circular String Pattern Matching Techniques

Similar lecture

Mining and Analysis of Air Quality to Aid Climate Change

An advanced deep learning model for short-term forecasting U.S. natural gas price and movement

.css-70qvj9{display:-webkit-box;display:-webkit-flex;display:-ms-flexbox;display:flex;-webkit-align-items:center;-webkit-box-align:center;-ms-flex-align:center;align-items:center;}Downloads

Next from AIAI 2020

Fingerprints Recognition System-Based on Mobile Device Identification using Circular String Pattern Matching Techniques

Similar lecture

Mining and Analysis of Air Quality to Aid Climate Change

Downloads