Nikolaos Passalis

Entropy Optimized Feature-Based Bag-of-Words Representation for Information Retrieval

In this paper, we present a supervised dictionary learning method for optimizing the feature-based Bag-of-Words (BoW) representation towards Information Retrieval. Following the cluster hypothesis, which states that points in the same cluster are likely to fulfill the same information need, we propose the use of an entropy-based optimization criterion that is better suited for retrieval instead of classification. We demonstrate the ability of the proposed method, abbreviated as EO-BoW, to improve the retrieval performance by providing extensive experiments on two multi-class image datasets. The BoW model can be applied to other domains as well, so we also evaluate our approach using a collection of 45 time-series datasets, a text dataset, and a video dataset. The gains are three-fold since the EO-BoW can improve the mean Average Precision, while reducing the encoding time and the database storage requirements. Finally, we provide evidence that the EO-BoW maintains its representation ability even when used to retrieve objects from classes that were not seen during the training.

Neural Bag-of-Features learning

In this paper, a neural learning architecture for the well-known Bag-of-Features (BoF) model, called Neural Bag-of-Features, is proposed. The Neural BoF model is formulated in two neural layers: a Radial Basis Function (RBF) layer and an accumulation layer. The ability of the Neural BoF model to improve the classification performance is demonstrated using four datasets, including a large-scale dataset, and five different feature types. The gains are two-fold: the classification accuracy increases and, at the same time, smaller networks can be used, reducing the required training and testing time. Furthermore, the Neural BoF natively supports training and classifying from feature streams. This allows the proposed method to efficiently scale to large datasets. The streaming process can also be used to introduce noise and reduce the over-fitting of the network. Finally, the Neural BoF provides a framework that can model and extend the dictionary learning methodology. HighlightsA neural generalization of the Bag-of-Features (BoF) model is introduced.The proposed model supports the discriminant weighting of the feature space.Two incremental algorithms (for training and classification) are proposed.A method for providing visual attention information for the BoF model is introduced.The proposed method is evaluated using four datasets from different domains.

Forecasting Stock Prices from the Limit Order Book Using Convolutional Neural Networks

In todays financial markets, where most trades are performed in their entirety by electronic means and the largest fraction of them is completely automated, an opportunity has risen from analyzing this vast amount of transactions. Since all the transactions are recorded in great detail, investors can analyze all the generated data and detect repeated patterns of the price movements. Being able to detect them in advance, allows them to take profitable positions or avoid anomalous events in the financial markets. In this work we proposed a deep learning methodology, based on Convolutional Neural Networks (CNNs), that predicts the price movements of stocks, using as input large-scale, high-frequency time-series derived from the order book of financial exchanges. The dataset that we use contains more than 4 million limit order events and our comparison with other methods, like Multilayer Neural Networks and Support Vector Machines, shows that CNNs are better suited for this kind of task.

Information Clustering Using Manifold-Based Optimization of the Bag-of-Features Representation

In this paper, a manifold-based dictionary learning method for the bag-of-features (BoF) representation optimized toward information clustering is proposed. First, the spectral representation, which unwraps the manifolds of the data and provides better clustering solutions, is formed. Then, a new dictionary is learned in order to make the histogram space, i.e., the space where the BoF historgrams exist, as similar as possible to the spectral space. The ability of the proposed method to improve the clustering solutions is demonstrated using a wide range of datasets: two image datasets, the 15-scene dataset and the Corel image dataset, one video dataset, the KTH dataset, and one text dataset, the RT-2k dataset. The proposed method improves both the internal and the external clustering criteria for two different clustering algorithms: 1) the

Spectral Clustering using Optimized Bag-of-Features

{k}

Learning Neural Bag-of-Features for Large-Scale Image Retrieval

-means and 2) the spectral clustering. Also, the optimized histogram space can be used to directly assign a new object to its cluster, instead of using the spectral space (which requires reapplying the spectral clustering algorithm or using incremental spectral clustering techniques). Finally, the learned representation is also evaluated using an information retrieval setup and it is demonstrated that improves the retrieval precision over the baseline BoF representation.

MultiSpot: Spotting Sentiments with Semantic Aware Multilevel Cascaded Analysis

In this paper a dictionary learning method for the Bag-of-Features (BoF) representation optimized towards spectral clustering is proposed. First, an objective function that measures the clustering ability of the histogram space, i.e., the space where the extracted histogram vectors lie after being encoded using the dictionary, is defined using the similarity graph of the spectral representation of the data. Then, the learned dictionary is optimized in order to minimize this objective and achieve better clustering solutions. That way, the histogram space is used as an intermediate space that helps to unwrap the manifold of the data. The ability of the proposed method to improve the spectral clustering is demonstrated using several clustering criteria and two image datasets, the 15-scene dataset and the Corel image dataset. The better clustering ability of the learned representation is also confirmed by evaluating the clustering solutions in the histogram space, as well as by using a different clustering algorithm. Although only image datasets were used in this work, the proposed method can be applied to any kind of objects that are represented using the BoF model, such as video, audio and time-series.

Dimensionality Reduction Using Similarity-Induced Embeddings

In this paper, the well-known bag-of-features (BoFs) model is generalized and formulated as a neural network that is composed of three layers: 1) a radial basis function (RBF) layer; 2) an accumulation layer; and 3) a fully connected layer. This formulation allows for decoupling the representation size from the number of used codewords, as well as for better modeling the feature distribution using a separate trainable scaling parameter for each RBF neuron. The resulting network, called retrieval-oriented neural BoF (RN-BoF), is trained using regular back propagation and allows for fast extraction of compact image representations. It is demonstrated that the RN-BoF model is capable of: 1) increasing the object encoding and retrieval speed; 2) reducing the extracted representation size; and 3) increasing the retrieval precision. A symmetry-aware spatial segmentation technique is also proposed to further reduce the encoding time and the storage requirements and allows the method to efficiently scale to large datasets. The proposed method is evaluated and compared to other state-of-the-art techniques using five different image datasets, including the large-scale YouTube Faces database.

Spatial Bag of Features Learning for Large Scale Face Image Retrieval

Given a textual resource (e.g. post, review, comment), how can we spot the expressed sentiment? What will be the core information to be used for accurately capturing sentiment given a number of textual resources? Here, we introduce an approach for extracting and aggregating information from different text-levels, namely words and sentences, in an effort to improve the capturing of documents’ sentiments in relation to the state of the art approaches. Our main contributions are: (a) the proposal of two semantic aware approaches for enhancing the cascaded phase of a sentiment analysis process; and (b) MultiSpot, a multilevel sentiment analysis approach which combines word and sentence level features. We present experiments on two real-world datasets containing movie reviews.

Improving Face Pose Estimation Using Long-Term Temporal Averaging for Stochastic Optimization

The vast majority of dimensionality reduction (DR) techniques rely on the second-order statistics to define their optimization objective. Even though this provides adequate results in most cases, it comes with several shortcomings. The methods require carefully designed regularizers and they are usually prone to outliers. In this paper, a new DR framework that can directly model the target distribution using the notion of similarity instead of distance is introduced. The proposed framework, called similarity embedding framework (SEF), can overcome the aforementioned limitations and provides a conceptually simpler way to express optimization targets similar to existing DR techniques. Deriving a new DR technique using the SEF becomes simply a matter of choosing an appropriate target similarity matrix. A variety of classical tasks, such as performing supervised DR and providing out-of-sample extensions, as well as, new novel techniques, such as providing fast linear embeddings for complex techniques, are demonstrated in this paper using the proposed framework. Six data sets from a diverse range of domains are used to evaluate the proposed method and it is demonstrated that it can outperform many existing DR techniques.