Shoji Mizobuchi

Suspended particulate matter sampling at an urban AERONET site in Japan, part 1: clustering analysis of aerosols

The aerosol properties of urban atmospheric particles have been analyzed for radiometry data obtained using a multi-spectral photometer located at a NASA/AERONET station at Kinki University Campus, Higashi-Osaka, Japan, since 2002. The suspended particulate matter has been simultaneously measured at the same AERONET site since 2004. Our measurements and a clustering analysis reveal the aerosol types over the industrial city of Higashi-Osaka. It is shown that aerosols at Higashi-Osaka can be classified into three categories: (1) ordinary, which represents the background clear atmosphere of Higashi-Osaka, (2) anthropogenic aerosol events, and (2) dust episodes called Kosa, when large amounts of soil dust are transported to Higashi-Osaka from the Chinese mainland on westerly winds, especially in spring. We also found that the linear correlation exists between column aerosol optical thickness and PM2.5 concentration, and the correlation is better within each aerosol cluster than overall.

A method for understanding time expressions

This paper proposes a method for understanding time expressions. For components of which time expressions consist, 31 concepts are defined. We provide formal representations for these concepts and proposed an algorithm to analyse the meaning of time expressions. The representation includes both conceptual and quantitative aspects. Experiments for about 2,000 time expressions extracted from actual documents show that the proposed method yields correct interpretation in about 93% of the cases.

Efficient multi-attribute pattern matching

This paper describes an efficient multi-attribute pattern matching machine to locate all occurrences of any of a finite number of a sequence of rule structures in a series of input structures. The matching operation of the proposed machine is similar to the method of Aho-Corasick or the method of retrieval using a trie, however, the proposed machine has the following distinctive features: (1) The proposed machine enables us to match set representations containing multiple attributes; (2) It enables us to match separate components; (3) It enables us to match a rule consisting of an exclusive set. In this paper, their features are described in detail. Moreover, the pattern matching algorithm is evaluated by the theoretical observations and the experimental observations that are supported by the simulation results for a variety of rules for document processing as text proofreading, text reduction, and examining a relation between sentences.

Conceptual and Quantitative Representations of Time Expressions

In natural language processing systems, it is very important to understand the meaning of time expressions and convert them into formal representations which include quantitative values. This paper proposes a method for understanding time expressions. For components of which time expressions consist, 31 concepts are introduced. Formal representations for these concepts are defined and an algorithm of analyzing the meaning of time expressions is proposed. The representation includes both conceptual and quantitative aspects in comparison with traditional approaches such as those of machine translation systems, and allows for more precise understanding of time expressions. An algorithm for translating time expressions into the corresponding formal representation by using a left-to-right parsing technique is presented. Experiments for about 2,000 time expressions extracted from actual documents show that the proposed method yields correct interpretation in about 93% of the cases.

An efficient representation for implementing finite state machines based on the double-array

Abstract This paper describes the double-array structure that is extended in order to apply it to general finite state machines. The double-array is an efficient data structure which combines time efficiency and space efficiency. However, the range of its application has been limited in the areas where a directed tree with labeled edges is used as the data structure. Here, the double-array structure is extended so that it can represent the graph structure and, furthermore, be operated dynamically. The presented method has been evaluated by theoretical observations, and its space efficiency is verified in our experiment.

An efficient retrieval algorithm for case structures using TRIE

Hierarchical classification is very comprehensive for human and a useful representation for many applications, such as selection of word translation of machine translation systems, semantic representation of natural language analysis, and so on. Especially, for the case structures, the case slot is generally restricted by hierarchical concepts. Case structures are a typical semantic representation for natural language understanding and it has semantic constraints, or roles, for noun phrases in the slot for a verb. Generally, semantic constraints are confirmed by hierarchical relationships between concepts of nouns and the input concepts. With growing hierarchical structures, the number of concepts increases and the length of numerical codes representing concepts becomes long. The frequency of determinations is very high for complex sentences. Multiple concepts correspond to one word. From these reasons, it costs a lot to determine whether a concept for a given word is a sub-concept for concepts of the case slot or not. This paper presents a fast method for determining the hierarchical relationships by introducing trie structures instead of a sequential storage of concept codes. The worst-case time complexity of the determination process by the presented method remarkably improves the one of the sequential storage, which depends on the number of concepts in the slot. The trie structures can be compressed by merging common transitions into one trie. From the simulation results, it is shown that the presented algorithm is 6 to 25 times faster than the sequential storage while keeping a smaller size of tries.

Web-based reading support system: assigning pronunciations to difficult words according to the vocabulary level of individual users

This paper proposes a system that helps a user read a document on the web more easily by assigning pronunciations of difficult words in the document based on the vocabulary level of the user and the difficulties of characters in the document. Vocabulary level of a user and the difficulties of characters were estimated using the Item Response Theory IRT. A simulation was conducted to evaluate the estimation of user vocabulary level. The results confirmed that the user vocabulary level can be estimated reasonably accurately when the user operates the pronunciations of half of the words with difficulties that are higher than his/her vocabulary level. The performance of the system was measured in terms of time and data size. The measurement results showed that the system performance was within an acceptable range.

Aerosol retrieval over urban cities

In order to understand the atmospheric constituents over the urban area, the aerosol properties obtained from radiometry with AERONET are compared with the suspended particulate matter (SPM) simultaneously measured with a new instrument (SPM-613D). It is found that the SPM measurements classified as fine particles (PM2.5) and coarse particles (TSP/PM10) are very useful for determining the dominating of particle size and air quality. There was a strong correlation between the PM2.5 concentration and aerosol index (AI). It indicates that aerosol characteristics can be estimated from SPM data, and vice versa.

Role of polarization in aerosol remote sensing

Acquisition of aerosol information on a global scale, especially over the continent, is an urgent task for the better understanding of Earths radiation budge and climate change. However, it is known that aerosol retrieval over land is much more difficult than that over ocean due to the complication of surface properties. A polarization sensor POLDER boarded on the satellite ADEOS-1 and -2 has shown that the polarization information is useful to extract the aerosol properties such as AOT (Aerosol Optical Thickness), and its wavelength tendency (Angstrom exponent) not only over the ocean but also over the land. Recently the polarization sensor becomes to be noticed. At such a time, this work focuses on role of polarization information from the point of view for aerosol retrieval. As a result, an improved map of aerosol is obtained based on POLDER-2 data. The obtained space based aerosols are examined with the radiometric measurements from the ground.

An efficient algorithm for retrieving example sentences

Storing and retrieving an useful example sentence efficiently is an important task in document management systems because text retrieval is the most time-consuming part of them. Inverted filing is a well-known approach, but there are some problems when storing a huge number of sentences. They arise when the intersection is computed for large postings indexed by terms, or keywords. Moreover, disk access for postings also takes a lot of time in this situation. This paper presents a technique for the storing of multi-stages of postings and retrieving them partly in order to compute efficiently the intersection between postings for the requested terms. From the simulation results, it is shown that the presented algorithm is 3 to 88 times faster than the traditional approach.