Link prediction in social networks, such as friendship networks and coauthorship networks, has recently attracted a great deal of attention. There have been numerous attempts to address the problem of link prediction through diverse approaches. In the present paper, we focus on the temporal behavior of the link strength, particularly the relationship between the time stamps of interactions or links and the temporal behavior of link strength and how link strength affects future link evolution. Most previous studies have not sufficiently discussed either the impact of time stamps of the interactions or time stamps of the links on link evolution. The gap between the current time and the time stamps of the interactions or links is also important to link evolution. In the present paper, we introduce a new time-aware feature, referred to as time score, that captures the important aspects of time stamps of interactions and the temporality of the link strengths. We also analyze the effectiveness of time score with different parameter settings for different network data sets. The results of the analysis revealed that the time score was sensitive to different networks and different time measures. We applied time score to two social network data sets, namely, Facebook friendship network data set and a coauthorship network data set. The results revealed a significant improvement in predicting future links.

Keyword-based linked data information retrieval is an easy choice for general-purpose users, but the implementation of such an approach is a challenge because mere keywords do not hold semantic information. Some studies have incorporated templates in an effort to bridge this gap, but most such approaches have proven ineffective because of inefficient template management. Because linked data can be presented in a structured format, we can assume that the data's internal statistics can be used to effectively influence template management. In this work, we explore the use of this influence for template creation, ranking, and scaling. Then, we demonstrate how our proposal for automatic linked data information retrieval can be used alongside familiar keyword-based information retrieval methods, and can also be incorporated alongside other techniques, such as ontology inclusion and sophisticated matching, in order to achieve increased levels of performance.

These days, the Web contains a huge volume of (semi-)structured data, called Linked Data (LD). However, LD suffer in data quality, and this poor data quality brings the need to identify erroneous data. Because manual erroneous data checking is impractical, automatic erroneous data detection is necessary. According to the data publishing guidelines of LD, data should use (already defined) ontology which populates type-annotated LD. Usually, the data type annotation helps in understanding the data. However, in our observation, the data type annotation could be used to identify erroneous data. Therefore, to automatically identify possible erroneous data over the type-annotated LD, we propose a framework that uses a novel nearest-neighbor based error detection technique. We conduct experiments of our framework on DBpedia, a type-annotated LD dataset, and found that our framework shows better performance of error detection in comparison with state-of-the-art framework.

The usefulness and usability of existing knowledge graphs (KGs) are mostly limited because of the incompleteness of knowledge compared to the growing number of facts about the real world. Most existing ontology-based KG completion methods are based on the closed-world assumption, where KGs are fixed. In these methods, entities and relations are defined, and new entity information cannot be easily added. In contrast, in open-world assumptions, entities and relations are not previously defined. Thus there is a vast scope to find new entity information. Despite this, knowledge acquisition under the open-world assumption is challenging because most available knowledge is in a noisy unstructured text format. Nevertheless, Open Information Extraction (OpenIE) systems can extract triples, namely (head text; relation text; tail text), from raw text without any prespecified vocabulary. Such triples contain noisy information that is not essential for KGs. Therefore, to use such triples for the KG completion task, it is necessary to identify competent triples for KGs from the extracted triple set. Here, competent triples are the triples that can contribute to add new information to the existing KGs. In this paper, we propose the Competent Triple Identification (CTID) model for KGs. We also propose two types of feature, namely syntax- and semantic-based features, to identify competent triples from a triple set extracted by a state-of-the-art OpenIE system. We investigate both types of feature and test their effectiveness. It is found that the performance of the proposed features is about 20% better compared to that of the ReVerb system in identifying competent triples.

Advances in deep reinforcement learning have demonstrated its effectiveness in a wide variety of domains. Deep neural networks are capable of approximating value functions and policies in complex environments. However, deep neural networks inherit a number of drawbacks. Lack of interpretability limits their usability in many safety-critical real-world scenarios. Moreover, they rely on huge amounts of data to learn efficiently. This may be suitable in simulated tasks, but restricts their use to many real-world applications. Finally, their generalization capability is low, the ability to determine that a situation is similar to one encountered previously. We present a method to combine external knowledge and interpretable reinforcement learning. We derive a rule-based variant version of the Sarsa(λ) algorithm, which we call Sarsa-rb(λ), that augments data with prior knowledge and exploits similarities among states. We demonstrate that our approach leverages small amounts of prior knowledge to significantly accelerate the learning in multiple domains such as trading or visual navigation. The resulting agent provides substantial gains in training speed and performance over deep q-learning (DQN), deep deterministic policy gradients (DDPG), and improves stability over proximal policy optimization (PPO).

Knowledge graphs (KG) play a crucial role in many modern applications. However, constructing a KG from natural language text is challenging due to the complex structure of the text. Recently, many approaches have been proposed to transform natural language text to triples to obtain KGs. Such approaches have not yet provided efficient results for mapping extracted elements of triples, especially the predicate, to their equivalent elements in a KG. Predicate mapping is essential because it can reduce the heterogeneity of the data and increase the searchability over a KG. In this article, we propose T2KG, an automatic KG creation framework for natural language text, to more effectively map natural language text to predicates. In our framework, a hybrid combination of a rule-based approach and a similarity-based approach is presented for mapping a predicate to its corresponding predicate in a KG. Based on experimental results, the hybrid approach can identify more similar predicate pairs than a baseline method in the predicate mapping task. An experiment on KG creation is also conducted to investigate the performance of the T2KG. The experimental results show that the T2KG also outperforms the baseline in KG creation. Although KG creation is conducted in open domains, in which prior knowledge is not provided, the T2KG still achieves an F1 score of approximately 50% when generating triples in the KG creation task. In addition, an empirical study on knowledge population using various text sources is conducted, and the results indicate the T2KG could be used to obtain knowledge that is not currently available from DBpedia.

Mapping instances to the Linked Open Data (LOD) cloud plays an important role for enriching information of instances, since the LOD cloud contains abundant amounts of interlinked instances describing the instances. Consequently, many techniques have been introduced for mapping instances to a LOD data set; however, most of them merely focus on tackling with the problem of heterogeneity. Unfortunately, the problem of the large number of LOD data sets has yet to be addressed. Owing to the number of LOD data sets, mapping an instance to a LOD data set is not sufficient because an identical instance might not exist in that data set. In this article, we therefore introduce a heuristic expansion based framework for mapping instances to LOD data sets. The key idea of the framework is to gradually expand the search space from one data set to another data set in order to discover identical instances. In experiments, the framework could successfully map instances to the LOD data sets by increasing the coverage to 90.36%. Experimental results also indicate that the heuristic function in the framework could efficiently limit the expansion space to a reasonable space. Based upon the limited expansion space, the framework could effectively reduce the number of candidate pairs to 9.73% of the baseline without affecting any performances.

Knowledge graphs have been shown to be useful to many tasks in artificial intelligence. Triples of knowledge graphs are traditionally structured by human editors or extracted from semi-structured information; however, editing is expensive, and semi-structured information is not common. On the other hand, most such information is stored as text. Hence, it is necessary to develop a method that can extract knowledge from texts and then construct or populate a knowledge graph; this has been attempted in various ways. Currently, there are two approaches to constructing a knowledge graph. One is open information extraction (Open IE), and the other is knowledge graph embedding; however, neither is without problems. Stanford Open IE, the current best such system, requires labeled sentences as training data, and knowledge graph embedding systems require numerous triples. Recently, distributed representations of words have become a hot topic in the field of natural language processing, since this approach does not require labeled data for training. These require only plain text, but Mikolov showed that it can perform well with the word analogy task, answering questions such as, “a is to b as c is to __?.” This can be considered as a knowledge extraction task from a text for finding the missing entity of a triple. However, the accuracy is not sufficiently high when applied in a straightforward manner to relations in knowledge graphs, since the method uses only one triple as a positive example. In this paper, we analyze why distributed representations perform such tasks well; we also propose a new method for extracting knowledge from texts that requires much less annotated data. Experiments show that the proposed method achieves considerable improvement compared with the baseline; in particular, the improvement in HITS@10 was more than doubled for some relations.

Autonomous vehicles and advanced driver assistant systems (ADAS) are receiving notable attention as research fields in both academia and private industry. Some decision-making systems use sets of logical rules to map knowledge of the ego-vehicle and its environment into actions the ego-vehicle should take. However, such rulesets can be difficult to create — for example by manually writing them — due to the complexity of traffic as an operating environment. Furthermore, the building blocks of the rules must be defined. One common solution to this is using an ontology specifically aimed at describing traffic concepts and their hierarchy. These ontologies must have a certain expressive power to enable construction of useful rules. We propose a process of generating sets of explanatory rules for ADAS applications from data using ontology as a base vocabulary and present a ruleset generated as a result of our experiments that is correct for the scope of the experiment.

This paper presents an ontology-based driving decision making system, which can promptly make safety decisions in real-world driving. Analyzing sensor data for improving autonomous driving safety has become one of the most promising issues in the autonomous vehicles research field. However, representing the sensor data in a machine understandable format for further knowledge processing still remains a challenging problem. In this paper, we introduce ontologies designed for autonomous vehicles and ontology-based knowledge base, which are used for representing knowledge of maps, driving paths, and perceived driving environments. Advanced Driver Assistance Systems (ADAS) are developed to improve safety of autonomous vehicles by accessing to the ontology-based knowledge base. The ontologies can be reused and extended for constructing knowledge base for autonomous vehicles as well as for implementing different types of ADAS such as decision making system.

Link prediction in social networks, such as friendship networks and coauthorship networks, has recently attracted a great deal of attention. There have been numerous attempts to address the problem of link prediction through diverse approaches. In the present paper, we focused on predicting links in social networks using information flow via active links. The information flow heavily depends on link activeness. The links become active if the interactions happen frequently and recently with respect to the current time. The time stamps of the interactions or links provide vital information for determining the activeness of the links. In the present paper, we introduced a new algorithm, referred to as T_Flow, that captures the important aspects of information flow via active links in social networks. We tested T_Flow with two social network data sets, namely, a data set extracted from Facebook friendship network and a coauthorship network data set extracted from ePrint archives. We compare the link prediction performances of T_Flow with the previous method PropFlow. The results of T_Flow method revealed a notable improvement in link prediction for facebook data and significant improvement in link prediction for coauthorship data.

Since it is difficult to understand or predict research trends, we proposed methodologies for understanding and predicting research trends in the sciences, focusing on the structures of grants in the Japan Society for the Promotion of Science (JSPS), a Japanese funding agency. Grant applications are suitable for predicting research trends because these are research plans for the future, different from papers, which report research outcomes in the past. We investigated research trends in science focusing on research histories identified in grant application data of JSPS. Then we proposed a model for predicting research trends, assuming that breakthrough research encourages researchers to change from their current research field to an entirely new research field. Using breakthrough research, we aim to obtain higher precision in the prediction results. In our experimental results, we found that research fields in Informatics correlate well with actual scientific research trends. We also demonstrated that our prediction models are effective in actively interacting research areas, which include Informatics and Social Sciences.

An ability to assess similarity lies close to the core of cognition. Its understanding support the comprehension of human success in tasks like problem solving, categorization, memory retrieval, inductive reasoning, etc, and this is the main reason that it is a common research topic. In this paper, we introduce the idea of semantic differences and commonalities between words to the similarity computation process. Five new semantic similarity metrics are obtained after applying this scheme to traditional WordNet-based measures. We also combine the node based similarity measures with a corpus-independent way of computing the information content. In an experimental evaluation of our approach on two standard word pairs datasets, four of the measures outperformed their classical version, while the other performed as well as their unmodified counterparts.

The Linking Open Data (LOD) cloud is a collection of linked Resource Description Framework (RDF) data with over 31 billion RDF triples. Accessing linked data is a challenging task because each data set in the LOD cloud has a specific ontology schema, and familiarity with the ontology schema used is required in order to query various linked data sets. However, manually checking each data set is time-consuming, especially when many data sets from various domains are used. This difficulty can be overcome without user interaction by using an automatic method that integrates different ontology schema. In this paper, we propose a Mid-Ontology learning approach that can automatically construct a simple ontology, linking related ontology predicates (class or property) in different data sets. Our Mid-Ontology learning approach consists of three main phases: data collection, predicate grouping, and Mid-Ontology construction. Experiments show that our Mid-Ontology learning approach successfully integrates diverse ontology schema with a high quality, and effectively retrieves related information with the constructed Mid-Ontology.

Linked data entity resolution is the detection of instances that reside in different repositories but co-describe the same topic. The quality of the resolution result depends on the appropriateness of the configuration, including the selected matching properties and the similarity measures. Because such configuration details are currently set differently across domains and repositories, a general resolution approach for every repository is necessary. In this paper, we present cLink, a system that can perform entity resolution on any input effectively by using a learning algorithm to find the optimal configuration. Experiments show that cLink achieves high performance even when being given only a small amount of training data. cLink also outperforms recent systems, including the ones that use the supervised learning approach.