As virtual home assistants are becoming more popular, there is an emerging need for supporting languages other than English. While more wide-spread or popular languages such as Spanish, French or Hindi are already integrated into existing home assistants like Google Home or Alexa, integration of other less-known languages such as Romanian is still missing. This paper explores the problem of Natural Language Understanding (NLU) applied to a Romanian home assistant. We propose a customized capsule neural network architecture that performs intent detection and slot filling in a joint manner and we evaluate how well it handles utterances containing various levels of complexity. The capsule network model shows a significant improvement in intent detection when compared to models built using the well-known Rasa NLU tool. Through error analysis, we observe clear error patterns that occur systematically. Variability in language when expressing one intent proves to be the biggest challenge encountered by the model.
1 November 2024 in Knowledge and Information Systems
Clustering algorithms are essential in data analysis, but evaluating their performance is challenging when the true labels are not available, especially for non-convex clusters. Traditional performance evaluation metrics struggle to identify clustering quality, often assigning higher scores for linearly separated clusters than the true clusters. We propose an original approach to distance computation that accounts for the data structure, thus improving the clustering quality evaluation for non-convex clusters without affecting other shapes of clusters. We also showcase the applicability of this method through a modified version of K-Means using the proposed method that is capable of correctly separating non-convex clusters. The validation included the analysis of performance and time complexity of 3 traditional clustering quality evaluation metrics and the K-Means clustering algorithm against their augmented versions with the proposed approach. This analysis conducted on 7 benchmark synthetic datasets and 6 real datasets with various numbers of examples and features of diverse characteristics and joint complexities: simple convex clusters, overlapped and imbalanced clusters, and non-convex clusters. Through these analyses, we show the ineffectiveness of traditional methods and that the proposed approach overcomes the weaknesses of traditional methods.
Space Breakdown Method (SBM) is a clustering algorithm that was developed specifically for low-dimensional neuronal spike sorting. Cluster overlap and imbalance are common characteristics of neuronal data that produce difficulties for clustering methods. SBM is able to identify overlapping clusters through its design of cluster centre identification and the expansion of these centres. SBM’s approach is to divide the distribution of values of each feature into chunks of equal size. In each of these chunks, the number of points is counted and based on this number the centres of clusters are found and expanded. SBM has been shown to be a contender for other well-known clustering algorithms especially for the particular case of two dimensions while being too computationally expensive for high-dimensional data. Here, we present two main improvements to the original algorithm in order to increase its ability to deal with high-dimensional data while preserving its performance: the initial array structure was substituted with a graph structure and the number of partitions has been made feature-dependent, denominating this improved version as the Improved Space Breakdown Method (ISBM). In addition, we propose a clustering validation metric that does not punish overclustering and such obtains more suitable evaluations of clustering for spike sorting. Extracellular data recorded from the brain is unlabelled, therefore we have chosen simulated neural data, to which we have the ground truth, to evaluate more accurately the performance. Evaluations conducted on synthetic data indicate that the proposed improvements reduce the space and time complexity of the original algorithm, while simultaneously leading to an increased performance on neural data when compared with other state-of-the-art algorithms.Code available at: https://github.com/ArdeleanRichard/Space-Breakdown-Method.
Neurons in the brain fire action potentials or spikes to encode information and communicate with each other. The firing can be in tonic mode, where individual spikes are emitted at relatively distant time intervals, and in burst mode, where neurons typically fire trains of spikes in short temporal succession. Functionally, neuronal bursts may be involved in reliable information transmission or specialized information encoding schemes. Understanding the dynamics of neuronal burst activity plays a crucial role in unravelling the complex mechanisms underlying neural information processing. However, detecting bursts in extracellularly recorded neural data is far from trivial. There are many methods for burst detection, but none was adopted universally, and they are based exclusively on the evaluation of spike times. This article aims to compare these burst detection algorithms against each other from several perspectives in order to establish the most suitable and robust one. Moreover, we propose a novel burst detection method considering additional features that contain as much information as possible from the recording.