Search

As electric vehicle (EV) emerges, it is important to understand how driver's driving behavior is influencing power consumption in an electric vehicle. Driver's personal driving behavior is usually quite distinctive and can be recognized by means of driving patterns after some driving cycles. This paper presents a method combining several machine learning approaches to characterize driving behaviors of electric vehicles. The driving patterns are modeled according to power consumption monitored by the battery management system (BMS), in aspects of individual driver's personal and EV-fleet operations. First, we apply an unsupervised clustering approach to characterize a driver's behaviors by formulating driving patterns. Subsequently, the resulting clustered datasets were used to train machine-learning based classifiers for classification of dataset of EV and EV-fleet driving patterns. The work aims to provide a robust solution to help identify the characteristics of specific types of EVs and their driver behaviors, in order to allow automakers and EV-subsystem providers to gather valuable driving information for product improvement.

Introduction Predicting the sentiments and emotions of people from their texts is a critical issue in cognitive computing. The explosive growth of social network services has led to a tremendous increase of textual data, increasing the demand of the advanced analysis of these data. Sentiment analysis on textual social media data emerged in recent years to fulfill the needs of areas such as national security, business, politics, and economics; however, text messages from social networks are rather different from those of traditional text documents, especially in presentation style and lengths. Therefore, it is difficult but essential to develop an effective method to explore the sentiments of social messages. Methods In this study, we first applied a self-organizing map (SOM) algorithm to cluster social messages as well as sentiment keywords. An association discovery process was then applied to discover the associations between a message and some sentiment keywords, and the sentiment of a message was determined according to such associations. Results We performed experiments on collected Twitter messages and the results’ accuracy outperformed that of a similar approach. Conclusions A sentiment analysis approach based on SOMs was proposed. The associations between messages and keywords were derived using the proposed method. The novelty of this work arises from the adoption of association discovery process in sentiment analysis.

This study aims to extract symptom profiles and functional impairments of major depressive disorder from electronic health records (EHRs). A chart review was conducted by three annotators on 500 discharge notes randomly selected from a medical center in Taiwan to compile annotated corpora for nine depressive symptoms and four types of functional impairment. Named entity recognition techniques including the dictionary-based approach., a conditional random field model, and deep learning approaches were developed for the task of recognizing depressive symptoms and functional impairments from EHRs. The results show that the average micro-F-measures of the supervised learning approaches in extracting depressive symptoms is almost perfect (>0.90) but less accurate for the extraction of functional impairment.

Graphical abstractThis work proposes a scheme to improve the self-organizing map algorithm. This is the overall flowchart of the proposed algorithm. The key ingredients include:1.A novel topic identification scheme.2.Lateral expansion using novel topic incompatibility measure.3.Hierarchical expansion scheme using novel cluster size and topic size criteria.Display Omitted HighlightsIncorporation of topic identification into SOM learning could be beneficial to text categorization task.Both lateral and hierarchical expansion during SOM learning were achieved according to criteria based on identified topics.The produced text hierarchies outperformed contemporary approaches in quality and performance on text categorization. Self-organizing maps (SOM) have been applied on numerous data clustering and visualization tasks and received much attention on their success. One major shortage of classical SOM learning algorithm is the necessity of predefined map topology. Furthermore, hierarchical relationships among data are also difficult to be found. Several approaches have been devised to conquer these deficiencies. In this work, we propose a novel SOM learning algorithm which incorporates several text mining techniques in expanding the map both laterally and hierarchically. On training a set of text documents, the proposed algorithm will first cluster them using classical SOM algorithm. We then identify the topics of each cluster. These topics are then used to evaluate the criteria on expanding the map. The major characteristic of the proposed approach is to combine the learning process with text mining process and makes it suitable for automatic organization of text documents. We applied the algorithm on the Reuters-21578 dataset in text clustering and categorization tasks. Our method outperforms two comparing models in hierarchy quality according to users' evaluation. It also receives better F1-scores than two other models in text categorization task.

In this paper, we describe our work on developing a model and method for extracting key entities from the online social messages regarding emergent events for enhancing ontology engineering, enabling a sensible solution for prevention of similar disasters. Our work started with the development of an event modelling system using a data-cluster slicing approach, which combines analytics of social data and event lifecycle algorithms, allowing for large-scale emerging novel events to be quickly and accurately analyzed. Subsequently, our system computes the energy of each collected event data sets, and then encapsulates ranked temporal, spatial and topical keywords into a structured node for event-entity extraction, in order to update event ontologies for fast response of emergent events. The preliminary experimental results demonstrate that our developed system is workable, allowing for prediction of possible evolution and early warning of critical incidents with a support of dynamic entity extraction.

In this paper, we address a big-data analysis method for estimating the driving range of an electric vehicle (EV), allowing drivers to overcome range anxiety. First, we present an estimating approach to project the life of battery pack for 1600 cycles (i.e., 8 years/160 000 km) based on the data collected from a cycle-life test. This approach has the merit of simplicity. In addition, it considers several critical issues that occur inside battery packs, such as the dependence of internal resistance and the state-of-health. Subsequently, we describe our work on driving pattern analysis of an EV, using a machine-learning approach, namely growing hierarchical self-organizing maps, to cluster the collected EV big data. This paper contains the analysis of energy consumption and driving range estimation for EVs, including powertrain simulation and driving behavior analysis. The experimental results, including both simulating battery degradation and analysis of driving behaviors, demonstrate a feasible solution for improving driving range estimation by the EV big data.

A common question around the development of electric vehicles (EVs) is associated with the power, performance and reliability of their battery-powered systems in real world driving situations. The aim of this research work is to evaluate and characterize an EV battery power under real-world driving conditions in order to inform the design of the next generation systems. Also, we develop a framework of Connected Electric Vehicle Cloud (also known as connected EV cloud, or EV-cloud) to collect energy patterns and analyze driving behaviors for EV energy management. In this work we used a machine learning method based on Google's TensorFlow framework (TensorSOM) as our kernel analytic tool. Additionally, we utilized the SOM Toolbox on the Matlab platform to confirm the TensorSOM clustered results. The experimental result demonstrated that our proposed approach is a sensible solution for learning EV drivers' behaviors.

With the consciousness of environmental protection is going to rising up globally, every government have seriously concerned the related issues of PM2.5. In essence, PM2.5 is the microscopic contaminated particles. The sizes of particles are just one of twenty-eighth of human's hair, in addition to penetrating the respiration system, carrying the heavy mental, dioxin and bacteria, through the blood circulation could intrude easily into the human alveoli. Furthermore, the contaminated particles may cause great damages to human health. Thus, in this work we investigate some factors which may influence the air quality related to PM2.5, including temperature, humidity, and wind speed. In particular, we collected the data of the factors in several cities in Taiwan for experimentation. After removing the outlier data samples in the database, we make use of the Support Vector Machines (SVM) and Support Vector Regression (SVR) techniques to experiment with the collected data for correlation analysis of regional air quality.

Social bookmarking Websites are popular nowadays for they provide platforms that are easy and clear to browse and organize Web pages. Users can add tags on Web pages to allow easy comprehension and retrieval of Web pages. However, tag spams could also be added to promote the opportunity of being referenced of a Web page, which is troublesome to users for accessing uninterested Web pages. In this work, we proposed a scheme to automatically detect such tag spams using a proposed text mining approach based on self-organizing map (SOM) model. We used SOM to find the associations among Web pages as well as tags. Such associations were then used to discover the relationships between Web pages and tags. Tag spams can then be detected according to such relationships. Experiments were conducted on a set of Web pages collected from a social bookmarking site and obtained promising result.

In an era of growing concern over climate change, several utility companies originally supplied wholesale and retail power mainly made by burning coal, have started to consider and build the clean-energy power systems for resolving global warming problems. Wind power is nowadays regarded as one of the predominant alternative sources of clean energy. In this paper, we discuss our work on utilizing climate big-data associated with wind power, collected from several wind farms over four years, for exploring the correlation between regional wind speed and wind power. Once this huge amount of data are analyzed, it can be used to develop policies for siting wind-power facilities, designing smart charging algorithms, or evaluating the capacity of electrical distribution systems to meet the actual requirement of power load. Our work started with collecting related climate data for building data model to perform analytics work and experiments using Support Vector Regression (SVR) method. Also, we observed the correlations between other factors related to wind speed and wind energy from our empirical model. The preliminary experimental results demonstrate that our developed system framework is workable, allowing for detailed analysis of the important wind-power related factors on specific wind farm regions.

Although diverse groups argue about the potential and true value benefits from social-media big data, there is no doubt that the era of big data exploitation has begun, driving the development of novel data-centric applications. Big data is notable not only because of its size, but also because of the complexity caused by its relationality to other data. In the past, owing to the limited possibilities of accessing big data, few data sources were available to allow researchers to develop advanced data-driven applications, such as monitoring of emerging real-world events. In fact, social media is greatly impacting the growth of big data; and big data is providing enterprises with the data to help them understand how to better detect marketing demands. Microblogging is a social network service capable of aggregating messages to explore facts and unknown knowledge. Nowadays, people often attempt to search for trending news and hot topics in real time from microblogging messages to satisfy their information needs. Under such a circumstance, a real demand is to find a way to allow users to organize a large number of microblogging messages into understandable events. In this work, we attempt to tackle such challenges by developing an online text-stream clustering approach using a modified density-based clustering model with collected microblogging big data. The system kernel combines three technical components, including a dynamic term weighting scheme, a neighbourhood generation algorithm and an online density-based clustering technique. After acquiring detected event topics by the system, our system provides functions for recommending top-priority event information to assist people to effectively organize emerging event data through the developed topic ranking algorithm.

Due to the explosive growth of social-media applications, enhancing event-awareness by social mining has become extremely important. The contents of microblogs preserve valuable information associated with past disastrous events and stories. To learn the experiences from past events for tackling emerging real-world events, in this work we utilize the social-media messages to characterize real-world events through mining their contents and extracting essential features for relatedness analysis. On one hand, we established an online clustering approach on Twitter microblogs for detecting emerging events, and meanwhile we performed event relatedness evaluation using an unsupervised clustering approach. On the other hand, we developed a supervised learning model to create extensible measure metrics for offline evaluation of event relatedness. By means of supervised learning, our developed measure metrics are able to compute relatedness of various historical events, allowing the event impacts on specified domains to be quantitatively measured for event comparison. By combining the strengths of both methods, the experimental results showed that the combined framework in our system is sensible for discovering more unknown knowledge about event impacts and enhancing event awareness.

This paper presents a machine learning approach to analyze driving behaviors that allows a better understanding of electric vehicle. We implement a pattern recognition process to model the driving pattern according to the energy consumption for both a single driver and a fleet. The growing hierarchical self-organizing maps (GHSOM) is applied to learn driver's behaviors gradually, and the experimental results show that the driving behaviors could be recognized with the increase of driving cycles. Moreover, the proposed framework would enhance the understanding of driver's behaviors and also facilitate the EV system design, the big data analytics for IoV and the implementation of advanced driver assistant system (ADAS).

Highlights? We applied a density-based stream clustering method for mining Twitter data. ? The developed method can detect real-time and geospatial event features. ? Using the detection results can estimate the temporal and spatial impacts of events. ? Our method is well suited for awareness of large-scale events and risk management. Social networks have been regarded as a timely and cost-effective source of spatio-temporal information for many fields of application. However, while some research groups have successfully developed topic detection methods from the text streams for a while, and even some popular microblogging services such as Twitter did provide information of top trending topics for selection, it is still unable to fully support users for picking up all of the real-time event topics with a comprehensive spatio-temporal viewpoint to satisfy their information needs. This paper aims to investigate how microblogging social networks (i.e. Twitter) can be used as a reliable information source of emerging events by extracting their spatio-temporal features from the messages to enhance event awareness. In this work, we applied a density-based online clustering method for mining microblogging text streams, in order to obtain temporal and geospatial features of real-world events. By analyzing the events detected by our system, the temporal and spatial impacts of the emerging events can be estimated, for achieving the goals of situational awareness and risk management.

Highlights? We developed a fusion method to integrate text and photo types of location information. ? It combines image annotation and geographic text mining. ? Photos and texts are mapped to a unified semantic space for fusion. ? The approach can associate text and image types of documents with correct geographical zones. Due to the steady increase in the number of heterogeneous types of location information on the internet, it is hard to organize a complete overview of the geospatial information for the tasks of knowledge acquisition related to specific geographic locations. The text- and photo-types of geographical dataset contain numerous location data, such as location-based tourism information, therefore defining high dimensional spaces of attributes that are highly correlated. In this work, we utilized text- and photo-types of location information with a novel approach of information fusion that exploits effective image annotation and location based text-mining approaches to enhance identification of geographic location and spatial cognition. In this paper, we describe our feature extraction methods to annotating images, and utilizing text mining approach to analyze images and texts simultaneously, in order to carry out geospatial text mining and image classification tasks. Subsequently, photo-images and textual documents are projected to a unified feature space, in order to generate a co-constructed semantic space for information fusion. Also, we employed text mining approaches to classify documents into various categories based upon their geospatial features, with the aims to discovering relationships between documents and geographical zones. The experimental results show that the proposed method can effectively enhance the tasks of location based knowledge discovery.

With the increasing number of multilingual texts in the internet, multilingual text retrieval techniques have become an important research issue. However, the discovery of relationships between different languages remains an open problem. In this paper we propose a method, which applies the growing hierarchical self-organizing map (GHSOM) model, to discover knowledge from multilingual text documents. Multilingual parallel corpora were trained by the GHSOM to generate hierarchical feature maps. A discovery process is then applied on these maps to discover the relationships between documents of different languages. The relationships between keywords of different languages are also revealed. We conducted experiments on a set of Chinese—English bilingual parallel corpora to discover the relationships between documents of these languages. We also use such relationships to perform multilingual information retrieval tasks. The experimental results show that our multilingual text mining approach may capture conceptual relationships among documents as well as keywords written in different languages.

Traditional content-based image retrieval (CBIR) systems often fail to meet a user's need due to the 'semantic gap' between the extracted features of the systems and the user's query. The cause of the semantic gap is the failure of extracting real semantics from an image and the query. To extract semantics of images, however, is a difficult task. Most existing techniques apply some predefined semantic categories and assign the images to appropriate categories through some learning processes. Nevertheless, these techniques always need human intervention and rely on content-based features. In this paper we propose a novel approach to bridge the semantic gap which is the major deficiency of CBIR systems. We conquer the deficiency by extracting semantics of an image from the environmental texts around it. Since an image generally co-exists with accompanying texts in various formats, we may rely on such environmental texts to discover the semantics of the image. We apply a text mining process, which adopts the self-organizing map (SOM) learning algorithm as a kernel, on the environmental texts of an image to extract the semantic information from this image. Some implicit semantic information of the images can be discovered after the text mining process. We also define a semantic relevance measure to achieve the semantic-based image retrieval task. We performed experiments on a set of images which are collected from web pages and obtained promising results.

This paper presents an expansible machine learning approach applying the EV big data as the human sensor to extract driving behaviors and driving modes. A pattern recognition approach is proposed to model the driving pattern according to the energy consumption of an EV. The growing hierarchical self-organizing maps (GHSOM) is applied to learn driver's behaviors gradually in the offline process, and the clustered neurons are used as the training sets for implementing online classifiers based on support vector machine (SVM). This proposed framework would facilitate the understanding of driver's behaviors and help drivers overcome range anxiety.

In the manufacturing of semiconductor encapsulation, the production yield is one of critical issues concerned by all foundries. It is because that yield rate can directly affects the quality of the final product and the profitability. In this work we take the defect-products as samples and use machine learning techniques to classify the samples and verify the accuracy and feasibility of the experiment. We use Support Vector Machines (SVM) model to perform classification and compare the resulting accuracy with the results of Back Propagation Neural Network (BPN) model. Furthermore, we employ a statistical method namely Pearson product-moment correlation coefficient to identify the influential factors for production quality. The experimental result demonstrates that our hybrid method has great potentials for yield improvement in semiconductor manufacturing.

In this work, an "analytical data model of mosquito vector" was developed to perform analytical computation to the character of the dengue vectors. Our goal is to investigate a way to understand how the temporal trend of collected dataset correlates with the incidence dengue as identified by national health authorities. Based upon the mosquito-vector big data collections, we investigate how changes in some specific variables such as rainfall, temperature, and humidity can dramatically affect the population of mosquito vectors, in order to provide early warnings of dengue outbreaks. Thus, our system will collectively collect online sensing data of the variables and store them in a database, in order to combine the historical big data as training datasets for analytical computation. Also, the developed model is able to merge the experimental datasets with current hot-topic information which is relevant to mosquito vectors obtained from data of social sensors (i.e. social messages). The experimental data show that our system is of great potentials in providing early warnings of dengue outbreaks.

Growth of Electric vehicles (EV) starts to change the way that people transit. Several factors that affact the performance of EVs and environment including energy efficiency, safety, product durability, climate, geographical factor, infrastructure, and grid capacity need to be further investigated to cope with upcoming challenges. These issues mainly involve three fields including information technology, EV design and battery management. With the demand to allow EV-data to connect to clouds, the big data collected from EVs creates an unprecedented opportunity for developing novel ways for transportation and information exchange. In this work, we demonstrate the process of pattern extraction of EV related data based on a battery model and characteristics of EV systems. Furthermore, the proposed approach provides an energy management scheme for drivers to overcome "range anxiety". We utilized the EV system data which is critical to the energy consumption to discover patterns for long-term performance estimation. We formulated driver's behaviors by training the operating data collected from a real EV system with an unsupervised learning algorithm by the GHSOM neural network model. The experimental result shows that our approach has high potential to explore driver's behavioral patterns and estimate the driving range. The proposed framework can be appled to new EV design, intelligent transportation system (ITS), and big data analytics for the fields of internet of vehicle as well as urban computing.

The research on automatic hypertext construction emerges rapidly in the last decade because there exists a urgent need to translate the gigantic amount of legacy documents into web pages. Unlike traditional 'flat' texts, a hypertext contains a number of navigational hyperlinks that point to some related hypertexts or locations of the same hypertext. Traditionally, these hyperlinks were constructed by the creators of the web pages with or without the help of some authoring tools. However, the gigantic amount of documents produced each day prevent from such manual construction. Thus an automatic hypertext construction method is necessary for content providers to efficiently produce adequate information that can be used by web surfers. Although most of the web pages contain a number of non-textual data such as images, sounds, and video clips, text data still contribute the major part of information about the pages. Therefore, it is not surprising that most of automatic hypertext construction methods inherit from traditional information retrieval research. In this work, we will propose a new automatic hypertext construction method based on a text mining approach. Our method applies the self-organizing map algorithm to cluster some at text documents in a training corpus and generate two maps. We then use these maps to identify the sources and destinations of some important hyperlinks within these training documents. The constructed hyperlinks are then inserted into the training documents to translate them into hypertext form. Such translated documents will form the new corpus. Incoming documents can also be translated into hypertext form and added to the corpus through the same approach. Our method had been tested on a set of at text documents collected from a newswire site. Although we only use Chinese text documents, our approach can be applied to any documents that can be transformed to a set of index terms.

Recently research on text mining has attracted lots of attention from both industrial and academic fields. Text mining concerns of discovering unknown patterns or knowledge from a large text repository. The problem is not easy to tackle due to the semi-structured or even unstructured nature of those texts under consideration. Many approaches have been devised for mining various kinds of knowledge from texts. One important aspect of text mining is on automatic text categorization, which assigns a text document to some predefined category if the document falls into the theme of the category. Traditionally the categories are arranged in hierarchical manner to achieve effective searching and indexing as well as easy comprehension for human beings. The determination of category themes and their hierarchical structures were most done by human experts. In this work, we developed an approach to automatically generate category themes and reveal the hierarchical structure among them. We also used the generated structure to categorize text documents. The document collection was trained by a self-organizing map to form two feature maps. These maps were then analyzed to obtain the category themes and their structure. Although the test corpus contains documents written in Chinese, the proposed approach can be applied to documents written in any language and such documents can be transformed into a list of separated terms.

This paper describes our work on developing a language-independent technique for discovery of implicit knowledge from multilingual information sources. Text mining has been gaining popularity in the knowledge discovery field, particularity with the increasing availability of digital documents in various languages from all around the world. However, currently most text mining tools mainly focus only on processing monolingual documents (particularly English documents): little attention has been paid to apply the techniques to handle the documents in Asian languages, and further extend the mining algorithms to support the aspects of multilingual information sources. In this work, we attempt to develop a language-neutral method to tackle the linguistics difficulties in the text mining process. Using a variation of automatic clustering techniques, which apply a neural net approach, namely the Self-Organizing Maps (SOM), we have conducted several experiments to uncover associated documents based on a Chinese corpus, Chinese-English bilingual parallel corpora, and a hybrid Chinese-English corpus. The experiments show some interesting results and a couple of potential paths for future work in the field of multilingual information discovery. Besides, this work is expected to act as a starting point for exploring the impacts on linguistics issues with the machine-learning approach to mining sensible linguistics elements from multilingual text collections.

In recent years, the concept of human-in-the-loop has been utilized to support environment sensing. Along with IoT (Internet of Things) and wearable computing technologies, connecting people and devices to the internet provides a significant advantage for real-time emergency management. For development of context-aware applications, it is important to utilize higher-level semantic information, such as human activity, social emotions, and human behaviors for event monitoring. Therefore, human users may become part of sensor networks by using mobile devices and social media to report local information around them. In this work, we mainly focus on the use of social messages spreading by human users to model the real-world events, in order to incorporate human sensors into event contexts for situational awareness. First, our algorithm computes the energy of each collected event messages, and then encapsulates ranked temporal, spatial and topical keywords into a structured node, which could reinforce the alert collected from physical nodes. The experimental results show that the proposed approach is able to extract essential entities of events for incorporating human sensors into event contexts for event prevention and risk management.

In this paper, we describe our work on extracting entities from the online social messages regarding emergent events for ontology learning, which can contribute to a solution for quick response of emerging disastrous events. Our work started with the development of a real-time event detection system using a data-cluster slicing approach which combines social data analysis and early warning algorithms, allowing for quickly detecting emerging large-scale events from collected tweets. Subsequently, our system computes the energy of each collected event dataset, and then encapsulates ranked temporal, spatial and topical keywords into a structured node for event-entity extraction, in order to learn and update event ontologies for fast response of emergent events. The preliminary experimental results demonstrate that our developed system is workable, allowing for prediction of possible evolution for early warning of critical incidents with a dynamic ontology engineering.

Self-organizing map (SOM) algorithm has been applied widely in tasks such as data clustering and visualization. Two major deficiencies of classical SOM are the need of predefined map structure and the lack of hierarchy generation. Several approaches have been devised to tackle these deficiencies. One of our previous works, namely the topic-oriented self-organizing map (TOSOM), tries to remedy the classical SOM by combining topic identification process into the training phase. In this work, we will further expand the learning algorithm of TOSOM by incorporating user's constraints. Both structural and topical constraints which specified by the user could be used to guide the learning process. Preliminary experiments demonstrate improvements over previous algorithm on text categorization task.

Nowadays, crowdsourcing has emerged as a popular and important problem-solving approach. The major difference between crowdsourcing and traditional outsourcing lies on the people which tasks were outsourced. Those people involved in crowdsourcing are generally varied in knowledge, demographic properties, and number. Many applications and services have been developed to solve various types of tasks. However, these applications and services focus on providing platforms for outsourcing to the crowd. Little has been addressed so far on the management and usage of those information produced during the crowdsourcing process. Actually, as an emerging social network application and service, the data and social interactions created during crowdsourcing should carry important and valuable knowledge. This knowledge will develop various techniques for mining messages and information of crowdsourcing process. In this work, we address several approaches to discover useful knowledge from data created for and in crowdsourcing process. We hope the outcome of this research could help discovering usable knowledge from such emerging social network services and bring benefit in constructing crowdsourcing services.

In this work we utilize the social messages to construct an extensible event ontology model for learning the experiences and knowledge to cope with emerging real-world events. We develop a platform combining several text mining and social analysis algorithms to cooperate with our stream mining approach to detecting large-scale disastrous events from social messages, in order to achieve the aim of automatically constructing event ontology for emergency response First, we employ the developed event detection technique on Twitter social-messages to monitor the occurrence of emerging events, and record the development and evolution of detected events. Furthermore, we store the messages associated with the detected events in a repository. Through the developed algorithms for analyzing the content of social messages and ontology construction the event ontology can be established, allowing for developing relevant applications for prediction of possible evolution and impact evaluation of the events in the future immediately, in order to achieve the goals for early warning of disasters and risk management.

The self-organizing map (SOM) model is a well-known neural network model with wide spread of applications. The main characteristics of SOM are two-fold, namely dimension reduction and topology preservation. Using SOM, a high-dimensional data space will be mapped to some low-dimensional space. Meanwhile, the topological relations among data will be preserved. With such characteristics, the SOM was usually applied on data clustering and visualization tasks. One major shortage of classical SOM learning algorithm is the necessity of predefined map topology. Furthermore, hierarchical relationships among data are also difficult to be revealed. In this work, we propose a novel SOM learning algorithm which incorporates several text mining techniques in expanding the map both laterally and hierarchically that could discover the relationships among documents in both perspectives. The proposed algorithm will first cluster a set of training documents using classical SOM algorithm. We then identify the topics of each cluster and use them to evaluate the criteria on expanding the map. We applied the algorithm on medium-size datasets and obtained promising result.

Intelligence collection and analysis always play a major role in a company's growth. Traditional intelligence management process was most concealed and required massive human effort. It also has the disadvantages of rarity and danger. Therefore open source intelligence (OSINT) emerged as a major intelligence collection and analysis approach. Differing from traditional approach, the sources of OSINT are publicly accessible and have the properties of openness and massiveness which may result in disadvantages of inconsistency and lack of validation. For now, most of the OSINT processing is conducted manually which requires massive human effort and time cost. Automatic processing of OSINT is then unavoidable for modern applications. Although there exists software services to aid such automatic processing, the functionality and degree of automation are still immature and limited. In this work we developed an automatic processing approach for OSINT based on proposed text mining techniques. This approach may automatically identify interesting events from various aspects from which business could benefit. The major contribution of this work is that we have developed high-order mining techniques for OSINT, which will benefit domains like national security, personal knowledge management, business intelligence, e-learning, etc.

The “microblogs” archives keep a valuable collection of records for past disastrous events and related stories. To learn the experiences from past microblogging messages for coping with emerging real-world events, in this work we utilize the social-media messages to characterize events for relatedness analysis. First, we established an online clustering algorithm to detect burstiness on Twitter microblogs for monitoring large-scale events. Furthermore, we compared the features extracted from the on-going event with the ones of other past detected events. Through the developed algorithms for computing event relatedness the ranking of related events can be sorted, allowing for predicting possible evolution and impacts of the event. For performing event clustering and relatedness computation, in this work we have developed an online system platform that can immediately evaluate the significantly emergent events to achieve the goals of real-time discovery of disastrous information and early warning for crisis management.

Nowadays lots of information are stored as Web pages for ease of sharing and searching. It is rather impractical for users to browse such gigantic amount of Web pages to obtain target information. Therefore, users often count on search engines to retrieve information they needed. Most of the search engines rely on the contents of Web pages, which are often represented by sets of keywords, to decide their relatedness to a user's query. However, such scheme is often imprecise since there is always a semantic gap between the actual semantics of a Web page and its representation. Manual annotations by human beings will certainly help since such annotations should always reflect the true meaning of the Web pages. However, ordinary annotators may not annotate a Web page with appropriate tags and thus deteriorate the quality of tags as well as the retrieval result. In this work, we propose a novel scheme for recommending tags while a user intends to annotate a Web page. Our scheme first discover the relationships across Web pages and tags based their clustering result. We then recommend tags to users according to such relationships. We tested our work on RSDC 2008 dataset and obtained promising result.

Social networks have been regarded as a timely and cost-effective source of spatio-temporal information for many fields of application. However, while some research groups have successfully developed topic detection methods from the text streams for a while, and even some popular microblogging services such as Twitter did provide information of top trending topics for selection, it is still unable to fully support users pickup all of the real-time event topics with a comprehensive spatio-temporal viewpoint to satisfy their information needs. This paper aims to enhance the understanding on how social networks can be used as a reliable source of spatio-temporal information, by analyzing the temporal and spatial dynamics of Twitter activity. In this work, we developed several algorithms for mining microblogging text stream to obtain real-time and geospatial event information. The goal of our approach is to effectively detecting and grouping emerging topics by making use of real-time messages and geolocation data provided by social network services.

Social book marking Web sites have emerged recently for collecting and sharing of interesting Web sites among users. People can add Web pages to such sites as bookmarks and allow themselves as well as others to manipulate them. One of the key features of the social book marking sites is the ability of annotating a Web page when it is being bookmarked. The annotation usually contains a set of words or phrases, which are collectively known as tags, that could reveal the semantics of the annotated Web page. Efficient and effective search of Web pages can then be achieved via such tags. However, spam tags that are irrelevant to the content of Web pages often appear to deceive other users for malicious or commercial purposes. Various techniques have been devised to tackle such tag spam detection problem. Most of these techniques were able to detect a user that always annotate spam tags. However, finer levels of detection are seldom discussed. In this work, we will propose a method based on a text mining approach to discover the relations between Web pages and there tag posts. These relations are then used to compute the similarity between a Web page and its tag post to decide if it is a spam post. Preliminary experiments show that the accuracy of the post-level spam detection task is 83%.

With the continuously increasing needs of location information for users around the world, applications of geospatial information have gained a lot of attention in both research and commercial organizations. Extraction of semantics from the image content for geospatial information seeking and knowledge discovery has been thus becoming a critical process. Unfortunately, the available geographic images may be blurred, either too light or too dark. It is therefore often hard to extract geographic features directly from images. In this paper, we describe our developed methods in applying local scale-invariant features and bag-of-keypoints techniques to annotating images, in order to carry out image categorization and geographic knowledge discovery tasks. First, local scale-invariant features are extracted from geographic images as representative geographic features. Subsequently, the bag-of-keypoints methods are used to construct a visual vocabulary and generate feature vectors to support image categorization and annotation. The annotated images are classified by using geographic nouns. The experimental results show that the proposed approach is sensible and can effectively enhance the tasks of geographic knowledge discovery.

Online microblogging services such as Twitter allow users to post very short messages related to everything ranging from mundane daily life routines to breaking news events. This phenomenon has changed the way for information acquisition. In this paper, we present an instinctive method with a time-decay function which corresponds to the natural propagation of social networks for clustering real-time text streams collected from Twitter. Compared to most previous studies, we follow natural cascading behaviors of event lifecycle to develop a self-adaptive clustering model for online event detection. Also, we construct an expandable similarity matrix which is capable of evaluating microblogging posts with incomplete semantic features. Experimental results show that the proposed method is a sensible solution to monitoring momentous real-time events and utilizing the text streams to facilitate the management of social networking data.

One of the basic human needs is to exchange information and socialize with each other. Online microblogging services such as Twitter allow users to post very short messages related to everything ranging from mundane daily life routines to breaking news events. A key challenging issue of mining such social messages is how to analyze the real-time distributed messages and extract significant features of them in a dynamic environment. In this work, we propose a novel term weighting method, called BursT, using sliding window techniques for weighting message streams. The experimental results show that our weighting technique has an outstanding performance to reflect the shifts of concept drift. The result of this work can be extended to perform a periodic feature extraction, and also be able to integrate other sophisticated clustering methods to enhance the efficiency for real-time event mining in social networks.

Social tags are annotations for Web pages collaboratively added by users. It will be much easier to understand the meaning of Web pages and classify them according to their tags. The precision in retrieving Web pages may also increase using such tags. Nowadays social tags are mostly annotated manually by users via social bookmarking Web sites. Such manual annotation process may produce diverse, redundant, and inconsistent tags. Besides, many tags which are inconsistent with their annotated Web pages exist and deteriorate the feasibility of social tags. In this work we will develop an automatic scheme to discover the associations between Web pages and social tags and apply such associations on applications of social tag spam detection. We applied a text mining approach based on self-organizing maps to find the relationships between Web pages and social tags. The disadvantages of manual annotation will be remedied through such relationships. The discovered associations were then used to identify social tag spams. Preliminary experiments show that the quality and usability of social tags were improved through this method.

Self-organizing map (SOM) learning algorithm has been widely applied in solving various tasks in pattern recognition, machine learning, and data mining, etc. Recently, it has been used to cluster documents and produced reasonable results. Traditional SOM algorithm learns from data using a fixed map. Approaches have been proposed to allow adaptable map structure. In this work, we propose a novel SOM learning algorithm that can expand the map laterally and hierarchically. The adaption of the map structure is based on topics identified from the underlying document clusters. This approach is different from traditional approaches which are typically driven by the data density of clusters. Preliminary experiment result suggested that the proposed algorithm outperforms other similar approaches.

With the increasing needs of location information, applications of geospatial information have gained a lot of attention in both research and commercial organizations Extraction of geospatial knowledge from the information content has been thus becoming a important process Among theses applications, a typical example is to discover relationships between various geospatial texts/data and specific locations In this paper, we describe a location based text mining approach using Artificial Neural Networks (ANN) to classify texts into various categories based on their geospatial features, with the aims to discovering relationships between documents and zones First, the collected documents were mapped to corresponding zones by the Adaptive Affinity Propagation (Adaptive AP) clustering techniques, then we performed framed maximize zones by means of Fuzzy ARTMAP (FAM) and Support Vector Machines (SVM) methods, allowing the results of relationships between documents and zones to be well presented Eventually, we compared our experimental results with that of baseline models using Self-organizing maps (SOM) and Learning Vector Quantization (LVQ) methods The preliminary results show that our platform framework has potential for geospatial knowledge discovery.

In this paper, we describe a location based text mining approach to classify texts into various categories based on their geospatial features, with the aims to discovering relationships between documents and zones. We first mapped documents into corresponding zones by Adaptive Affinity Propagation (Adaptive AP) clustering technique, and then framed maximize zones by means of Simplified Fuzzy ARTMAP (SFAM) and Support Vector Machines (SVM) methods. Also, we compared our experimental results with the baseline approaches of Self-organizing maps (SOM) and Learning Vector Quantization (LVQ) methods. The preliminary results show that our platform framework has the potential for geospatial data mining.

Due to the improvement of digital image technology and increasing amount of digital image data, the issue of automatic image annotation technology and applications becomes more and more important. In order to retrieval image efficiently, it is important to extract and represent the semantics of images. Traditional image semantics extraction and representation schemes were commonly divided into two categories, namely visual features and text annotations. However, visual feature scheme are difficult to extract and are often semantically inconsistent. On the other hand, the image semantics can be well represented by text annotations. It is also easier to retrieve images according to their annotations. The problem is annotating images are always time-consuming and requiring lots of human effort. In this thesis, we try to develop an automatic image annotation method. We adopted the Growing Hierarchical Self-Organizing Map (GHSOM) to help us discover the concealed relations between image data and annotation data, and annotate image according to such relations. We first applied GHSOM to cluster and generate hierarchies for images and annotations individually. We then proposed a hierarchical mapping method to discover the relations between image clusters and annotation clusters. New images can be annotated according to such relations. We conducted experiments using the proposed method on an image dataset and obtained promising result, which showed that our method could be plausible.

In this paper we describe our work on developing a novel technique for discovery of implicit knowledge about patents from multilingual patent information sources. In this work we developed a system platform to support locating similar and relevant multilingual patent documents. The platform was implemented using a multilingual vector space based on the latent semantic indexing (LSI) model, and utilizing collected professional Chinese-English parallel corpora for training the system model. These multilingual patent documents could then be mapped into the semantic vector space for evaluating their similarity by means of text clustering techniques. The preliminary results show that our platform framework has potential for retrieval and relatedness evaluation of multilingual patent documents.

This paper describes our work on developing a language-independent technique for discovery of implicit knowledge about patents from multilingual patent information ources. Traditional techniques of multi- and cross-language patent retrieval are mostly based on the process of translation. One major problem of those is that it is difficult to find related patents produced from other countries in a stand-alone patent information system. In this paper, we present a novel system platform to support locating similar and relevant multilingual patent documents. The platform is developed using a multilingual vector space based on the latent semantic indexing (LSI) model, and utilizing collected professional Chinese-English parallel corpora for training the system model. These multilingual patent documents can then be mapped into the semantic vector space for evaluating their similarity by means of text clustering techniques. The preliminary results show that our platform framework has potential for retrieval and relatedness evaluation of multilingual patent documents.

Web directories cluster Web pages into categories and usually organize them into hierarchies. Many users used them to browse for interesting Web pages in a coarse-to-fine manner. Nowadays most of the Web directories access monolingual Web pages and provide only monolingual interface which may limit the coverage and accessibility of Web pages for users familiar only with their native languages. Bilingual or multilingual Web directories thus may relieve such limitations. In this work, we develop an automated process to create multilingual (or bilingual, specifically) Web directories from a set of parallel corpora. We adopted the self-organizing map model to cluster the Web pages and construct Web directories for each language. A hierarchy alignment process was then applied on these monolingual hierarchies to obtain the relationships between different languages. A multilingual Web directory was then created using such relationships. We conducted experiments on a set of parallel corpora and the result demonstrated that our method could be feasible.

In this paper, we describe a two-stage hybrid approach to select gene features and produce dominant patterns for evaluating the pathological probability. To discover suitable genes as experiment samples for distinguishing the status of gene regulation, we utilized Receiver Operating Characteristic (ROC) method to eliminate non-significant genes of unapparent variation between normal tissues and tumors. Subsequently, these selected genes are clustered through an unsupervised learning algorithm to reduce overall training samples under the same condition. In addition, the resulting samples have been verified by means of experimenting with the SVM and KNN methods. The experimental results show that our approach has potentials to effectively reduce samples for microarray analysis.

Multilingual information retrieval has attracted lots of attention in recent years due to the explosive increase of multilingual Web pages. It will not be easy to retrieve documents written in languages other than the query if the relationships among entities of different languages were not found. In this work, we will develop a method based on self-organizing maps to organize documents into hierarchy. When hierarchies of individual languages are developed, an alignment process is then applied on these hierarchies to discover the associations between entities of different languages. The method had been performed on a set of parallel corpora containing Chinese and English documents and obtained promising result.

The Web pages nowadays were written in various languages including English, Chinese, Spanish, etc. There are increasing needs in searching Web pages of different languages using single query. This task is called multilingual information retrieval (MLIR). However, MLIR is difficult to achieve since we need some kind of method to find the associations between linguistic elements of different languages. In this work, we provide a method based on GHSOM to discover the associations between different languages and apply this method on MLIR task. The experiments show that our method provide a promising approach to tackle MLIR task.

With the increasing amount of multilingual texts in the Internet, multilingual text retrieval techniques have become an important research issue. However, the discovery of relationships between different languages remains an open problem. In this paper we propose a method, which applies the growing hierarchical self- organizing map (GHSOM) model, to discover knowledge from multilingual text documents. Multilingual parallel corpora were trained by the GHSOM to generate hierarchical feature maps. A discovery process is then applied on these maps to discover the relationships between documents of different languages. The relationships between keywords of different languages are also revealed. We conducted experiments on a set of Chinese-English bilingual parallel corpora to discover the relationships between documents of these languages.