High Stability Anomaly Detection in Random Environments

We propose anomaly detection to refine input data for predictive machine learning systems. When training, if there are outliers such as spike noises mixed in the input data, the quality of the trained model is deteriorated. The removing such outliers would be expected the service quality of machine learning systems improves such as autonomous vehicles and ship navigation. Conventionally, anomaly detection methods generally require the support of domain experts, and they could not treat with unstable random environments well. We propose a new anomaly detection method, which is highly stable and is capable of treating with random environments without experts. The proposed methods focus on a pairwise correlation between two input time-series, change rates of them are calculated and summarized on a quadrant chart for further analysis. The experiment using an open time-series dataset shows that the proposed methods successfully detect anomalies, and the detected data points are easily illustrated in a human-interpretable way.

The Uncertainty of Counterfactuals in Deep Learning

Counterfactuals have become a useful tool for explainable Artificial Intelligence (XAI). Counterfactuals provide various perturbations to a data instance to yield an alternate classification from a machine learning model. Several algorithms have been designed to generate counterfactuals using deep neural networks; however, despite their growing use in many mission-critical fields, there has been no investigation to date as to the epistemic uncertainty of generated counterfactuals. This could result in the use of risk-prone explanations in these fields. In this work, we use several data sets to compare the epistemic uncertainty of original instances to that of counterfactuals generated from those instances. As part of our analysis, we also measure the extent to which counterfactuals can be considered anomalies in those data sets. We find that counterfactual uncertainty is higher in three of the four datasets tested. Moreover, our experiments suggest a possible connection between reconstruction error using a deep autoencoder and the difference in epistemic uncertainty between training data and counterfactuals generated from that training data for a deep neural network.

Population-Based Novelty Searches Can Converge

Novelty search is a powerful tool for finding sets of complex objects in complicated, open-ended spaces. Recent empirical analysis on a simplified version of novelty search makes it clear that novelty search happens at the level of the archive space, not the individual point space. The sparseness measure and archive update criterion create a process that is driven by a clear pair of objectives: spread out to cover the space, while trying to remain as efficiently packed as possible driving these simplified variants to converge to an

Analysis of comparative performance of deep learning models for sentiment analysis

Sentiment analysis of text can be performed using machine learning and natural language processing methods. However, there is no single tool or method that is effective in all cases. The objective of this research project is to determine the effectiveness of neural network-based architecture to perform sentiment analysis of customer comments and reviews, such as the ones on Amazon site. A typical sentiment analysis process involves text preparation (of acquired content), sentiment detection, sentiment classification and analysis of results. In this research, the objective is to a) identify the best approach for text preparation in a given application (text filtering approach to remove errors in data), and, most importantly, b) what is the best machine learning (feed forward neural nets, convolutional neural nets, Long Short-Term Memory networks) approach that provides best classification accuracy. In this research, a set of three thousand two hundred reviews of food related products were used to train and experiment with a neural network-based sentiment analysis system. The neural network implementation of six different models provided close to one-hundred percent accuracy of test data, and a decent test accuracy in mid-80%. The results of the research would be useful to businesses in evaluating customer preferences for products or services.

System Z for Conditional Belief Bases with Positive and Negative Information

In non-monotonic reasoning, conditional belief bases mostly contain positive information in the form of standard conditionals. However, in practice we are often confronted with negative information, stating that a conditional does \emph{not} hold, i.e. we need a suitable approach for reasoning over belief bases $\Delta$ with positive and negative information. In this paper, we investigate the interaction of positive and negative information in a conditional belief base and establish a property for partitions of $\Delta$ that is equivalent to consistency. Based on this property, we develop a non-trivial extension of system Z for mixed conditional belief bases and provide an algorithm to compute this partition.

Detecting the Presence of Named Entities in Bengali: Corpus and Experiments

Named Entity Recognition (NER) belongs to the field of Information Extraction (IE) and Natural LanguageProcessing (NLP). NER aims to find and categorize named entities present in the textual data into recognizable classes. Named entities play vital roles in other related fields like question-answering, relationship extraction, and machine translation. Researchers have done a significant amount of work (e.g., dataset construction and analysis) in this direction for several languages like English, Spanish, Chinese, Russian, Arabic, to name a few. We do not find a comparable amount of work for several South-Asian languages like Bengali/Bangla. Hence, as part of the initial phase, we have constructed a qualitative dataset in Bengali.In this paper, we identify the presence of Named Entities (NEs) in the Bengali text (sentences), classify them in standardized categories, and test whether an automatic detection of NE is possible. We present a new corpus and experimental results. Our dataset, annotated by multiple humans, shows promising results (F-measures ranging from 0.72 to 0.84) in different setups (support vector machine (SVM) setups with simple language features and Long-Short Term Memory (LSTM) setup with various word embedding).

On the Behaviour of Weighted Permutation Entropy on Fractional Brownian Motion in the Univariate and Multivariate Setting

The estimation of the qualitative behaviour of fractional Brownian motion is an important topic for modelling real-world applications. Permutation entropy is a well-known approach to quantify the complexity of univariate time series in a scalar-valued representation. As an extension often used for outlier detection, weighted permutation entropy takes amplitudes within time series into account. As many real-world problems deal with multivariate time series, these measures need to be extended though. First, we introduce multivariate weighted permutation entropy, which is consistent with standard multivariate extensions of permutation entropy. Second, we investigate the behaviour of weighted permutation entropy on both univariate and multivariate fractional Brownian motion and show revealing results.