Showing papers in "IEEE Intelligent Systems in 2020"

FedHealth: A Federated Transfer Learning Framework for Wearable Healthcare

Abstract: With the rapid development of computing technology, wearable devices make it easy to get access to people's health information. Smart healthcare achieves great success by training machine learning models on a large quantity of user personal data. However, there are two critical challenges. First, user data often exist in the form of isolated islands, making it difficult to perform aggregation without compromising privacy security. Second, the models trained on the cloud fail on personalization. In this article, we propose FedHealth, the first federated transfer learning framework for wearable healthcare to tackle these challenges. FedHealth performs data aggregation through federated learning, and then builds relatively personalized models by transfer learning. Wearable activity recognition experiments and real Parkinson's disease auxiliary diagnosis application have evaluated that FedHealth is able to achieve accurate and personalized healthcare without compromising privacy and security. FedHealth is general and extensible in many healthcare applications.

A Secure Federated Transfer Learning Framework

Abstract: Machine learning relies on the availability of vast amounts of data for training. However, in reality, data are mostly scattered across different organizations and cannot be easily integrated due to many legal and practical constraints. To address this important challenge in the field of machine learning, we introduce a new technique and framework, known as federated transfer learning (FTL), to improve statistical modeling under a data federation. FTL allows knowledge to be shared without compromising user privacy and enables complementary knowledge to be transferred across domains in a data federation, thereby enabling a target-domain party to build flexible and effective models by leveraging rich labels from a source domain. This framework requires minimal modifications to the existing model structure and provides the same level of accuracy as the nonprivacy-preserving transfer learning. It is flexible and can be effectively adapted to various secure multiparty machine learning tasks.

The Hourglass Model Revisited

Abstract: Recent developments in the field of AI have fostered multidisciplinary research in various disciplines, including computer science, linguistics, and psychology. Intelligence, in fact, is much more than just IQ: it comprises many other kinds of intelligence, including physical intelligence, cultural intelligence, linguistic intelligence, and emotional intelligence (EQ). While traditional classification tasks and standard phenomena in computer science are easy to define, however, emotions are still a rather mysterious subject of study. That is why so many different emotion classifications have been proposed in the literature and there is still no common agreement on a universal emotion categorization model. In this article, we revisit the Hourglass of Emotions, an emotion categorization model optimized for polarity detection, based on some recent empirical evidence in the context of sentiment analysis. This new model does not claim to offer the ultimate emotion categorization but it proves the most effective for the task of sentiment analysis.

Research on Road Traffic Situation Awareness System Based on Image Big Data

Abstract: Road traffic is an important component of the national economy and social life. Promoting intelligent and Informa ionization construction in the field of road traffic is conducive to the construction of smart cities and the formulation of macro strategies and construction plans for urban traffic development. Aiming at the shortcomings of the current road traffic system, this article, on the basis of combining convolution neural network, situational awareness technology, database and other technologies, takes the road traffic situational awareness system as the research object, and analyzes the information collection, processing, and analysis process of road traffic situational awareness system. Convolutional neural networks (CNN), region-CNN (R-CNN), fast R-CNN, and faster R-CNN are used for vehicle class classification and location identification in road image big data. The deep convolutional neural network model based on road traffic image big data was further established, and the system requirements analysis and system framework design and implementation were carried out. Through the analysis and trial of actual cases, the results show the application effect of the realized road traffic situational awareness system, which provides a scientific reference and basis for the establishment of modern intelligent transportation system.

A Sustainable Incentive Scheme for Federated Learning

Abstract: In federated learning (FL), a federation distributedly trains a collective machine learning model by leveraging privacy preserving technologies. However, FL participants need to incur some cost for contributing to the FL models. The training and commercialization of the models will take time. Thus, there will be delays before the federation could pay back the participants. This temporary mismatch between contributions and rewards has not been accounted for by existing payoff-sharing schemes. To address this limitation, we propose the FL incentivizer (FLI). It dynamically divides a given budget in a context-aware manner among data owners in a federation by jointly maximizing the collective utility while minimizing the inequality among the data owners, in terms of the payoff received and the waiting time for receiving payoffs. Comparisons with five state-of-the-art payoff-sharing schemes show that FLI attracts high-quality data owners and achieves the highest expected revenue for a federation.

A New Representation of Intuitionistic Fuzzy Systems and Their Applications in Critical Decision Making

Abstract: In decision-making problem, fuzzy system is considered as an effective tool with access to uncertain information by fuzzy representations. Evolutionary fuzzy systems have been developed with the appearance of intuitionistic fuzzy, hesitant fuzzy, neutrosophic representations, etc. Moreover, by capturing compound features and convey multifaceted information, complex numbers are utilized to generalize fuzzy and intuitionistic fuzzy sets. However, the order relations established in these existing systems have certain limitations, such as they are not total order relations or they are defined based on intermediate functions, hence it is difficult to use in building and ensuring important properties of logical operators and distance measures in the systems. In this article, a representation of the intuitionistic fuzzy systems based on complex numbers (IFS-C) in the polar form by a new way is proposed to overcome the above restrictions. Specifically, an intuitionistic fuzzy set is characterized by the two functions of modulus and argument. A new order relation, set-theoretic operations, and a new distance measure by the polar form of IFS-C are defined and investigated. The applicability of the proposal is illustrated by a new decision-making model called P-distance measure. It is tested on the benchmark medical datasets in comparison with the existing methods. The experiments confirm the advantages of the proposal.

Preserving User Privacy for Machine Learning: Local Differential Privacy or Federated Machine Learning?

Abstract: The growing number of mobile and IoT devices has nourished many intelligent applications. In order to produce high-quality machine learning models, they constantly access and collect rich personal data such as photos, browsing history, and text messages. However, direct access to personal data has raised increasing public concerns about privacy risks and security breaches. To address these concerns, there are two emerging solutions to privacy-preserving machine learning, namely local differential privacy and federated machine learning. The former is a distributed data collection strategy where each client perturbs data locally before submitting to the server, whereas the latter is a distributed machine learning strategy to train models on mobile devices locally and merge their output (e.g., parameter updates of a model) through a control protocol. In this article, we conduct a comparative study on the efficiency and privacy of both solutions. Our results show that in a standard population and domain setting, both can achieve an optimal misclassification rate lower than 20% and federated machine learning generally performs better at the cost of higher client CPU usage. Nonetheless, local differential privacy can benefit more from a larger client population ($>$> 1k). As for privacy guarantee, local differential privacy also has flexible control over the data leakage.

Decision Making in IoT Environment through Unsupervised Learning

Abstract: Nowadays, unsupervised learning can provide new perspectives to identify hidden patterns and classes inside the huge amount of data coming from the Internet of Things (IoT) world. Analyzing IoT data through machine learning techniques requires the use of mathematical algorithms, computational techniques, and an accurate tuning of the input parameters. In this article, we present a study of unsupervised learning techniques applied on IoT data to support decision-making processes inside intelligent environments. To assess the proposed approach, we discuss two case studies in which behavioral IoT data have been collected, also in a noninvasive way, in order to achieve an unsupervised classification that can be adopted during a decision-making process. The use of unsupervised learning techniques is acquiring a key role to complement the more traditional services with new decision-making ones supporting the needs of companies, stakeholders, and consumers.

XGBoost Model and Its Application to Personal Credit Evaluation

Abstract: This article investigates the application of the eXtreme Gradient Boosting (XGB) method to the credit evaluation problem based on big data. We first study the theoretical modeling of the credit classification problem using XGB algorithm, and then we apply the XGB model to the personal loan scenario based on the open data set from Lending Club Platform in USA. The empirical study shows that the XGB model has obvious advantages in both feature selection and classification performance compared to the logistic regression and the other three tree-based models.

Intent Classification for Dialogue Utterances

Abstract: In this work, we investigate several machine learning methods to tackle the problem of intent classification for dialogue utterances. We start with bag-of-words in combination with Naive Bayes. After that, we employ continuous bag-of-words coupled with support vector machines (SVM). Then, we follow long short-term memory (LSTM) networks, which are made bidirectional. The best performing model is hierarchical, such that it can take advantage of the natural taxonomy within classes. The main experiments are a comparison between these methods on an open sourced academic dataset. In the first experiment, we consider the full dataset. We also consider the given subsets of data separately, in order to compare our results with state-of-the-art vendor solutions. In general we find that the SVM models outperform the LSTM models. The former models achieve the highest macro-F1 for the full dataset, and in most of the individual datasets. We also found out that the incorporation of the hierarchical structure in the intents improves the performance.

Cross-Lingual Sentiment Quantification

Abstract: Sentiment Quantification is the task of estimating the relative frequency of sentiment-related classes—such as ${\sf Positive}$Positive and ${\sf Negative}$Negative—in a set of unlabeled documents. It is an important topic in sentiment analysis, as the study of sentiment-related quantities and trends across a population is often of higher interest than the analysis of individual instances. In this article, we propose a method for cross-lingual sentiment quantification, the task of performing sentiment quantification when training documents are available for a source language $\mathcal {S}$S, but not for the target language $\mathcal {T}$T, for which sentiment quantification needs to be performed. Cross-lingual sentiment quantification (and cross-lingual text quantification in general) has never been discussed before in the literature; we establish baseline results for the binary case by combining state-of-the-art quantification methods with methods capable of generating cross-lingual vectorial representations of the source and target documents involved. Experiments on publicly available datasets for cross-lingual sentiment classification show that the presented method performs cross-lingual sentiment quantification with high accuracy.

Three Algorithms for Solving High-Dimensional Fully Coupled FBSDEs Through Deep Learning

Abstract: Recently, the deep learning method has been used for solving forward–backward stochastic differential equations (FBSDEs) and parabolic partial differential equations, as it has good accuracy and performance for high-dimensional problems. In this article, we mainly solve fully coupled FBSDEs through deep learning and provide three algorithms, and the numerical results show remarkable performance, especially for high-dimensional cases.

Federated Generative Privacy

Abstract: We propose FedGP, a framework for privacy-preserving data release in the federated learning setting. We use generative adversarial networks, generator components of which are trained by FedAvg algorithm, to draw private artificial data samples and empirically assess the risk of information disclosure. Our experiments show that FedGP is able to generate labeled data of high quality to successfully train and validate supervised models. Finally, we demonstrate that our approach significantly reduces vulnerability of such models to model inversion attacks.

Formal Analysis of Smart Contract Based on Colored Petri Nets

Abstract: Smart contracts increasingly cause attention for its ability to widen blockchain's application scope. However, the security of contracts is vital to its wide deployment. In this article, we propose a multilevel smart contract modeling solution to analyze the security of contract. We improve the program logic rules for bytecode and apply the Hoare condition to create a Colored Petri Net (CPN) model. The model detection method provided by the CPN tools can show the full-state space and the wrong execution path, which help us analyze the security of the contract from several perspectives. The example shows that the counter-example path given by the contract model is accord with our expected results based on code analysis, proving the correctness of the solution. In addition, we design a highly automated modeling method, introducing custom call libraries and a path derivation algorithm based on backtracking, which improves the efficiency and pertinence of the dynamic simulation of CPN models.

Personalized Geographical Influence Modeling for POI Recommendation

Abstract: Point-of-interest (POI) recommendation has great significance in helping users find favorite places from a large number of candidate venues. One challenging in POI recommendation is to effectively exploit geographical information since users usually care about the physical distance to the recommended POIs. Though spatial relevance has been widely considered in recent recommendation methods, it is modeled only from the POI perspective, failing to capture user personalized preference to spatial distance. Moreover, these methods suffer from a diversity-deficiency problem since they are often based on collaborative filtering which always favors popular POIs. To overcome these problems, we propose in this article a personalized geographical influence modeling method called PGIM, which jointly learns users’ geographical preference and diversity preference for POI recommendation. Specifically, we model geographical preference from three aspects: user global tolerance, user local tolerance, and spatial distance. We also extract user diversity preference from interactions among users for diversity-promoting recommendation. Experimental results on three real-world datasets demonstrate the superiority of PGIM.

Commonsense Knowledge Enhanced Memory Network for Stance Classification

Abstract: Stance classification aims at identifying, in the text, the attitude toward the given targets as favorable, negative, or unrelated. In existing models for stance classification, only textual representation is leveraged, while commonsense knowledge is ignored. In order to better incorporate commonsense knowledge into stance classification, we propose a novel model named commonsense knowledge enhanced memory network, which jointly represents textual and commonsense knowledge representation of given target and text. The textual memory module in our model treats the textual representation as memory vectors, and uses attention mechanism to embody the important parts. For commonsense knowledge memory module, we jointly leverage the entity and relation embeddings learned by TransE model to take full advantage of constraints of the knowledge graph. Experimental results on the SemEval dataset show that the combination of the commonsense knowledge memory and textual memory can improve stance classification.

Proxy Experience Replay: Federated Distillation for Distributed Reinforcement Learning

Abstract: Traditional distributed deep reinforcement learning (RL) commonly relies on exchanging the experience replay memory (RM) of each agent. Since the RM contains all state observations and action policy history, it may incur huge communication overhead while violating the privacy of each agent. Alternatively, this article presents a communication-efficient and privacy-preserving distributed RL framework, coined federated reinforcement distillation (FRD). In FRD, each agent exchanges its proxy experience RM (ProxRM), in which policies are locally averaged with respect to proxy states clustering actual states. To provide FRD design insights, we present ablation studies on the impact of ProxRM structures, neural network architectures, and communication intervals. Furthermore, we propose an improved version of FRD, coined mixup augmented FRD (MixFRD), in which ProxRM is interpolated using the mixup data augmentation algorithm. Simulations in a Cartpole environment validate the effectiveness of MixFRD in reducing the variance of mission completion time and communication cost, compared to the benchmark schemes, vanilla FRD, federated RL (FRL), and policy distillation.

A Deep Coupled LSTM Approach for USD/CNY Exchange Rate Forecasting

Abstract: Forecasting CNY exchange rate accurately is a challenging task due to its complex coupling nature, which includes market-level coupling from interactions with multiple financial markets, macrolevel coupling from interactions with economic fundamentals, and deep coupling from interactions of the two aforementioned kinds of couplings. This study develops a new deep coupled long short-term memory (LSTM) approach, namely, DC-LSTM, to capture the complex couplings for USD/CNY exchange rate forecasting. In this approach, a deep structure consisting of stacked LSTMs is built to model the complex couplings. The experimental results with 10 years data indicate that the proposed approach significantly outperforms seven other benchmarks. The DC-LSTM is verified to be a useful tool to make wise investment decisions through a profitability discussion. The purpose in this article is to clarify the importance of coupling learning for exchange rate forecasting, and the usefulness of deep coupled model to capture the couplings.

A Novel Siamese-Based Approach for Scene Change Detection With Applications to Obstructed Routes in Hazardous Environments

Abstract: The demand for automatic scene change detection has massively increased in the last decades due to its importance regarding safety and security issues. Although deep learning techniques have provided significant enhancements in the field, such methods must learn which object belongs to the foreground or background beforehand. In this article, we propose an approach that employs siamese U-Nets to address the task of change detection, such that the model learns to perform semantic segmentation using background reference frames only. Therefore, any object that comes up into the scene defines a change. The experimental results show the robustness of the proposed model over the well-known public dataset CDNet2014. Additionally, we also consider a private dataset called “PetrobrasROUTES,” which comprises obstruction or abandoned objects in escape routes in hazardous environments. Moreover, the experiments show that the proposed approach is more robust to noise and illumination changes.

Battlefield Image Situational Awareness Application Based on Deep Learning

Abstract: With the rapid development of information technology, it has become an important topic to construct a situational awareness system that can independently mine data and information as well as perceive environmental situations by using deep learning. First, this article introduced the structure of convolutional neural networks (CNN) and You Only Look Once (YOLO) model. Then, it analyzed the structure and function of battlefield situational awareness system, and concluded that: in the whole situational awareness system, the discovery, category, and location analysis of situational elements, namely object target, is the foundation and key to realize the function. On this basis, this article establishes a battlefield situational awareness model based on the YOLO model. Finally, five common objects on the battlefield (helicopter gunship, missile, tank, soldier and gun) are classified and located, respectively. The YOLO model based on CNN is used to process the input image, and then the position, category, and corresponding confidence probability of all objects in the image are obtained directly, which realizes end-to-end learning, greatly improves the speed of target detection, and lays a foundation for assessing the battlefield situation.

Recognizing Nested Named Entity Based on the Neural Network Boundary Assembling Model

Abstract: The task to recognize named entities is often modeled as a sequence labeling process, which selects a label path whose probability is maximum for an input sentence. Because it makes the assumption that the input sentence has a flattened structure, it often fails to recognize nested named entities. In our previous work, a boundary assembling (BA) model was proposed. It is a cascading framework, which identifies named entity boundaries first, and then assembles them into entity candidates for further assessment. This model is effective to recognize nested named entities, but still suffers from poor performance caused by the sparse feature problem. In this article, the BA model is remodeled with the advancement of neural networks, which enables the model to capture semantic information of a sentence by using word embeddings pretrained in external resources. In our experiments, it shows an impressive improvement on the final performance, outperforming the state of the art more than 17% in F-score.

MDA: Multimodal Data Augmentation Framework for Boosting Performance on Image-Text Sentiment/Emotion Classification Tasks

Abstract: Multimodal data analysis has drawn increasing attention with the explosive growth of multimedia data. Although traditional unimodal data analysis tasks have accumulated abundant labeled datasets, there are few labeled multimodal datasets due to the difficulty and complexity of multimodal data annotation, nor is it easy to directly transfer unimodal knowledge into multimodal data. Unfortunately, there are only a little related data augmentation work in multimodal domain, especially for image-text data. In this paper, to address the scarcity problem of labeled multimodal data, we propose a Multimodal Data Augmentation (MDA) framework for boosting performance on multimodal classification task. Our framework learns a cross-modality matching network to select image-text pairs from existing unimodal datasets as the multimodal synthetic dataset, and uses this dataset to enhance the performance of classifiers. We take the multimodal sentiment analysis and multimodal emotion analysis as the experimental tasks and the experimental results show the effectiveness of our framework for boosting the performance on multimodal classification task.

A Data-Analytics Approach for Risk Evaluation in Peer-to-Peer Lending Platforms

Abstract: The goal of this article is to investigate the roles of individual behavior characteristics and Internet finance industry risk in the light of bank run theory for P2P. We know that risk evaluation is clearly important for peer-to-peer (P2P) lending platforms in China, as during the last two years, the industry has experienced thousands of platform crashes. Traditional approaches to evaluate enterprise risk are increasingly ineffective in this industry, due to the difficulty of assessing the real information. In addition, the Internet business model makes it possible to record new kinds of information. By applying a data-driven analytics method, we build an intelligent risk evaluation model for P2P platforms that have comparable targeting platforms. The case study shows that our risk evaluation method can generate early warning signals regarding platform or industry risk, which is able to provide effective supporting for P2P business in practice.

Blockchain-Based Fair and Secure Electronic Double Auction Protocol

Abstract: Double auction is an auction in which multiple buyers and sellers looking for a price where supply and demand balance. Since the electronic double auction based on secure multiparty computation (DABSMPC) cannot guarantee its fairness, we propose a blockchain-based fair and secure electronic double auction protocol (BFSDA). BFSDA modifies the data input and distribution mechanism of participants in DABSMPC, which improves the security of the protocol. Then, the BFSDA introduces and improves the blockchain-based fair and secure multiparty computation protocol (BFSMPC) to ensure fairness while increasing the success rate of secret recovery. In addition, BFSDA uses a fairer and more efficient protocol for secure two-party comparing to obtain the final marketing clearing price. The schema analysis result of BFSDA shows that: first, the private input data will not be revealed as long as data owner is not compromised, second, honest participants can get the result or economic compensation, and third, participants only need to pay the deposit once and a large amount of complicated verification operations are carried out off the chain, which ensures the efficiency of the protocol.

Image Polarity Detection on Resource-Constrained Devices

Abstract: Image polarity detection opens new vistas in the area of pervasive computing. State-of-the-art frameworks for polarity detection often prove computationally demanding, as they rely on deep learning networks. Thus, one faces major issues when targeting their implementation on resource-constrained embedded devices. This article presents a design strategy for convolutional neural networks that can support image-polarity detection on edge devices. The outcomes of experimental sessions, involving standard benchmarks and a pair of commercial edge devices, confirm the approach suitability.

Stock Selection Model Based on Machine Learning with Wisdom of Experts and Crowds

Abstract: Both stock recommendations from sell-side analysts and online user generated content from crowds have great significance in the stock market. We examine and compare different effects of analyst attitude and crowd sentiment on stock prices in this article with data from CSMAR. By estimating a multivariate linear regression model, we find that although the wisdom of both experts and crowds has impact on stock prices, the latter's impact on stock prices prevails. We also adopt LightGBM, a novel machine learning model, to predict stock trends based on empirical results. Portfolio returns of different models also suggest that crowd wisdom is more valuable for creating investment strategy than expert wisdom. And it is necessary to take the wisdom of both experts and crowds into consideration when making investment decision.

Data Cleaning for Personal Credit Scoring by Utilizing Social Media Data: An Empirical Study

Abstract: With the accumulation of data on personal behavior and the development of machine learning models and algorithms, it is becoming possible to use social media data for personal credit scoring. In this article, we use the systematic sampling method to obtain Douban's social media data. Because there are many abnormal users in these data, they are “real but false data” for personal credit evaluation. In order to better carry out personal credit scoring, we propose three criteria, power exponents of time interval distribution of individual user $\gamma _i$γi, user activity $A_i$Ai, and the ratio of out-degree and in-degree $R_i$Ri of user $i$i, which are used to systematically clean the data. And then, we used the logistic regression method to score the individual credits of users before and after data cleaning, and found that the rank order of personal credit scoring has changed significantly. This change is largely attributed to the changes of network structure after data cleaning. We believe that our work is very important to use the social media data to establish a credible personal credit evaluation system to reduce the credit risk of the current Internet financial industry.

MGNN: Mutualistic Graph Neural Network for Joint Friend and Item Recommendation

Abstract: Many social studies and practical cases suggest that people's consumption behaviors and social behaviors are not isolated but interrelated in social network services. However, most existing research either predicts users’ consumption preferences or recommends friends to users without dealing with them simultaneously. We propose a holistic approach to predict users’ preferences on friends and items jointly and thereby make better recommendations. To this end, we design a graph neural network that incorporates a mutualistic mechanism to model the mutual reinforcement relationship between users’ consumption behaviors and social behaviors. Our experiments on the two-real world datasets demonstrate the effectiveness of our approach in both social recommendation and link prediction.

Special Issue on Situation Awareness in Intelligent Human-Computer Interaction for Time Critical Decision Making

Abstract: The articles in this special section focus on situation awareness in intelligent human-computer interaction for critical decision making (HCI). HCI is recognized as an ctive field that focuses on the various interactions of human with machines. The HCI has been widely applied in multiple domains, such as artificial intelligence, computer vision, image and multimedia analysis, and cognitive and behavioral sciences. The objective of the HCI is to make the computer smart via receiving enough knowledge about the environment where it is deployed and reduce the human intervention aspect toward decision making. This enables development of high-end computers that are context aware and smart in making decisions with reference to the context. Situation awareness of an intelligent HCI will decide the success and application of the solution across the real world environment. The aim of this special issue is to provide a platform on the topic of situation awareness in intelligent HCI for time critical decision making.

Crowd Intelligence: Conducting Asymmetric Impact-Performance Analysis Based on Online Reviews

Abstract: Asymmetric impact-performance analysis (AIPA) is an effective technique for understanding customer satisfaction and formulating improvement strategies for products and services. Typically, AIPA is conducted based on data obtained from customer surveys, which are expensive in terms of time and money. As a new data source, online reviews have many advantages, which are a promising data source for conducting AIPA. To this end, this article proposes a method for conducting AIPA based on online reviews. To illustrate the feasibility and validity of the proposed method, a case study of AIPA for a five-star hotel in Singapore is given. The proposed method can provide managers one more choice for conducting AIPA with lower cost and shorter time since products and services online reviews can be easily collected.