Many machine learning algorithms require the input to be represented as a fixed-length feature vector. When it comes to texts, one of the most common fixed-length features is bag-of-words. Despite their popularity, bag-of-words features have two major weaknesses: they lose the ordering of the words and they also ignore semantics of the words. For example, "powerful," "strong" and "Paris" are equally distant. In this paper, we propose Paragraph Vector, an unsupervised algorithm that learns fixed-length feature representations from variable-length pieces of texts, such as sentences, paragraphs, and documents. Our algorithm represents each document by a dense vector which is trained to predict words in the document. Its construction gives our algorithm the potential to overcome the weaknesses of bag-of-words models. Empirical results show that Paragraph Vectors outperforms bag-of-words models as well as other techniques for text representations. Finally, we achieve new state-of-the-art results on several text classification and sentiment analysis tasks.

/pdf/distributed-representations-of-sentences-and-documents-3g8596uykm.pdf

Distributed Representations of Sentences and Documents

Semantic word spaces have been very useful but cannot express the meaning of longer phrases in a principled way. Further progress towards understanding compositionality in tasks such as sentiment detection requires richer supervised training and evaluation resources and more powerful models of composition. To remedy this, we introduce a Sentiment Treebank. It includes fine grained sentiment labels for 215,154 phrases in the parse trees of 11,855 sentences and presents new challenges for sentiment compositionality. To address them, we introduce the Recursive Neural Tensor Network. When trained on the new treebank, this model outperforms all previous methods on several metrics. It pushes the state of the art in single sentence positive/negative classification from 80% up to 85.4%. The accuracy of predicting fine-grained sentiment labels for all phrases reaches 80.7%, an improvement of 9.7% over bag of features baselines. Lastly, it is the only model that can accurately capture the effects of negation and its scope at various tree levels for both positive and negative phrases.

/pdf/recursive-deep-models-for-semantic-compositionality-over-a-35h7g7u52l.pdf

Recursive Deep Models for Semantic Compositionality Over a Sentiment Treebank

The ability to accurately represent sentences is central to language understanding. We describe a convolutional architecture dubbed the Dynamic Convolutional Neural Network (DCNN) that we adopt for the semantic modelling of sentences. The network uses Dynamic k-Max Pooling, a global pooling operation over linear sequences. The network handles input sentences of varying length and induces a feature graph over the sentence that is capable of explicitly capturing short and long-range relations. The network does not rely on a parse tree and is easily applicable to any language. We test the DCNN in four experiments: small scale binary and multi-class sentiment prediction, six-way question classification and Twitter sentiment prediction by distant supervision. The network achieves excellent performance in the first three tasks and a greater than 25% error reduction in the last task with respect to the strongest baseline.

/pdf/a-convolutional-neural-network-for-modelling-sentences-3vzv81a42n.pdf

A Convolutional Neural Network for Modelling Sentences

Many machine learning algorithms require the input to be represented as a fixed-length feature vector. When it comes to texts, one of the most common fixed-length features is bag-of-words. Despite their popularity, bag-of-words features have two major weaknesses: they lose the ordering of the words and they also ignore semantics of the words. For example, "powerful," "strong" and "Paris" are equally distant. In this paper, we propose Paragraph Vector, an unsupervised algorithm that learns fixed-length feature representations from variable-length pieces of texts, such as sentences, paragraphs, and documents. Our algorithm represents each document by a dense vector which is trained to predict words in the document. Its construction gives our algorithm the potential to overcome the weaknesses of bag-of-words models. Empirical results show that Paragraph Vectors outperform bag-of-words models as well as other techniques for text representations. Finally, we achieve new state-of-the-art results on several text classification and sentiment analysis tasks.

Leu-M

The rapid technology evolution and the future challenges demand us to adopt increasingly cutting-edge and up-to-date technological tools. In this perspective, Blockchain technology is revolutionizing countless areas of our daily lives. In this paper, we present a non exhaustive list on the state of the art about Blockchain technology in multiple application fields, both from an industry and business perspective and from a consumer one. We will also present a dedicated focus on the frictions between distributed ledgers and data protection regulations crossing all these areas.

Blockchain, State-of-the-Art and Future Trends

The improvement of the Quality of Life (QoL) and the enhancement of the Quality of Services (QoS) represent the main goal of every city evolutionary process. It is possible making cities smarter promoting innovative solutions by use of Information and Communication Technology (ICT) for collecting and analysing large amounts of data generated by several sources, such as sensor networks, wearable devices, and IoT devices spread among the city. The integration of different technologies and different IT systems, needed to build smart city applications and services, remains the most challenge to overcome. In the Smart City context, this paper intends to investigate the Smart Environment pillar, and in particular the aspect related to the implementation of Smart Energy Grid for citizens in the urban context. The innovative characteristic of the proposed solution consists of using the Blockchain technology to join the Grid, exchanging information, and buy/sell energy between the involved nodes (energy providers and private citizens), using the Blockchain granting ledger.

/pdf/smarter-city-smart-energy-grid-based-on-blockchain-399dq3adb6.pdf

Smarter City: Smart Energy Grid based on Blockchain Technology

In distributional semantics studies, there is a growing attention in compositionally determining the distributional meaning of word sequences. Yet, compositional distributional models depend on a large set of parameters that have not been explored. In this paper we propose a novel approach to estimate parameters for a class of compositional distributional models: the additive models. Our approach leverages on two main ideas. Firstly, a novel idea for extracting compositional distributional semantics examples. Secondly, an estimation method based on regression models for multiple dependent variables. Experiments demonstrate that our approach outperforms existing methods for determining a good model for compositional distributional semantics.

/pdf/estimating-linear-models-for-compositional-distributional-1yoetyl98q.pdf

Estimating Linear Models for Compositional Distributional Semantics

There are several areas in which organisations can adopt technologies that will support decision-making: artificial intelligence is one of the most innovative technologies that is widely used to assist organisations in business strategies, organisational aspects and people management. In recent years, attention has increasingly been paid to human resources (HR), since worker quality and skills represent a growth factor and a real competitive advantage for companies. After having been introduced to sales and marketing departments, artificial intelligence is also starting to guide employee-related decisions within HR management. The purpose is to support decisions that are based not on subjective aspects but on objective data analysis. The goal of this work is to analyse how objective factors influence employee attrition, in order to identify the main causes that contribute to a worker’s decision to leave a company, and to be able to predict whether a particular employee will leave the company. After the training, the obtained model for the prediction of employees’ attrition is tested on a real dataset provided by IBM analytics, which includes 35 features and about 1500 samples. Results are expressed in terms of classical metrics and the algorithm that produced the best results for the available dataset is the Gaussian Naive Bayes classifier. It reveals the best recall rate (0.54), since it measures the ability of a classifier to find all the positive instances and achieves an overall false negative rate equal to 4.5% of the total observations.

Predicting Employee Attrition Using Machine Learning Techniques

The Internet of Things (IoT) devices are able to collect and share data directly with other devices through the cloud environment, providing a huge amount of information to be gathered, stored and analyzed for data-analytics processes. The scenarios in which the IoT devices may be useful are amazing varying, from automotive, to industrial automation or remote monitoring of domestic environment. Furthermore, has been proved that healthcare applications represent an important field of interest for IoT devices, due to the capability of improving the access to care, reducing the cost of healthcare and most importantly increasing the quality of life of the patients. In this paper, we analyze the state-of-art of IoT in medical environment, illustrating an extended range of IoT-driven healthcare applications that, however, still need innovative and high technology-based solutions to be considered ready to market. In particular, problems regarding characteristics of response-time and precision will be examined.  Furthermore, wearable and energy saving properties will be investigated in this paper and also the IT architectures able to ensure security and privacy during the all data-transmission process. Finally, considerations about data mining applications, such as risks prediction, classification and clustering will be provided, that are considered fundamental issues to ensure the accuracy of the care processes.

/pdf/e-health-iot-universe-a-review-1r1sooo6r2.pdf

Francesca Fallucchi

Papers

Blockchain, State-of-the-Art and Future Trends

Smarter City: Smart Energy Grid based on Blockchain Technology

Estimating Linear Models for Compositional Distributional Semantics

Predicting Employee Attrition Using Machine Learning Techniques

E-health-IoT Universe: A Review