Although there has been a great deal of research on automatic summarization, most methods rely on statistical methods, disregarding relationships between extracted textual segments. We propose a novel method to extract a set of comprehensible sentences which centers on several key points to ensure sentence connectivity. It features a similarity network from documents with a lexical dictionary, and spreading activation to rank sentences. We show evaluation results of a multi-document summarization system based on the method participating in a competition of summarization, TSC (Text Summarization Challenge) task, organized by the third NTCIR project.

The recent increase in the use of intelligent devices such as smartphones has enhanced the relationship between daily human behavior sensing and useful applications in ubiquitous computing. This paper proposes a novel method inspired by personal sensing technologies for collecting and visualizing road accessibility at lower cost than traditional data collection methods. To evaluate the methodology, we recorded outdoor activities of nine wheelchair users for approximately one hour each by using an accelerometer on an iPod touch and a camcorder, gathered the supervised data from the video by hand, and estimated the wheelchair actions as a measure of street level accessibility in Tokyo. The system detected curb climbing, moving on tactile indicators, moving on slopes, and stopping, with F-scores of 0.63, 0.65, 0.50, and 0.91, respectively. In addition, we conducted experiments with an artificially limited number of training data to investigate the number of samples required to estimate the target.

Two types of similarities between words have been studied in the natural language processing community: synonymy and relational similarity. A high degree of similarity exist between synonymous words. On the other hand, a high degree of relational similarity exists between analogous word pairs. We present and empirically test a hypothesis that links these two types of similarities. Specifically, we propose a method to measure the degree of synonymy between two words using relational similarity between word pairs as a proxy. Given two words, first, we represent the semantic relations that hold between those words using lexical patterns. We use a sequential pattern clustering algorithm to identify different lexical patterns that represent the same semantic relation. Second, we compute the degree of synonymy between two words using an inter-cluster covariance matrix. We compare the proposed method for measuring the degree of synonymy against previously proposed methods on the Miller-Charles dataset and the WordSimilarity-353 dataset. Our proposed method outperforms all existing Web-based similarity measures, achieving a statistically significant Pearson correlation coefficient of 0.867 on the Miller-Charles dataset.

This paper discusses crystal-growth and device-design issues associated with the development of high-performance InP/InGaAs heretostructure bipolar transistors (HBTs). It is shown that a highly Si-doped n+-subcollector in the HBT structure causes anomalous Zn redistribution during metalorganic vapor phase epitaxial (MOVPE) growth. A thermodynamical model of and a useful solution to this big problem are presented. A novel hybrid structure consisting of an abrupt emitter-base heterojunction and a compositionally-graded base is shown to enhance nonequilibrium base transport and thereby increase current gain and cutoff frequency fT. A double-heterostructure bipolar transistor (DHBT) with a step-graded InGaAsP collector can improve collector breakdown behavior without any speed penalty. We also elucidate the effect of emitter size shrinkage on high-frequency performance. Maximum oscillation frequency fmax in excess of 250 GHz is reported.

This paper describes IC-oriented high-performance AlGaAs/GaAs heterojunction bipolar transistors that were fabricated to demonstrate their great potential in applications to high-speed integrated circuits. A collector structure of ballistic collection transistors with a launcher (LBCTs) shortens the intrinsic delay time of the transistors. A novel and simple self-aligned fabrication process, which features an base-metal-overlaid structure (BMO), reduces emitter- and base-resistances and collector capacitance. The combination of the thin-collector LBCT layer structure and the BMO self-alignment technology raises the average value of cutoff frequency, fT, to 160 GHz with a standard deviation as small as 4.3 GHz. By modifying collector thickness and using Pt/Ti/Pt/Au as the base ohmic contact metal in BMO-LBCTs, the maximum oscillation frequency, fmax, reaches 148 GHz with a 114 GHz fT. A 2:1 multiplexer with retiming D-type flip-flops (DFFs) at input/output stages fabricated on a wafer with the thin-collector LBCT structure operates at 19 Gbit/s. A monolithic preamplifier fabricated on the same wafer has a transimpedance of 52 dBΩ with a 3-dB-down bandwidth of 18.5 GHz and a gain S21 OF 21 dB with a 3-dB-down bandwidth of 19 GHz. Finally, a 40 Gbit/s selector IC and a 50 GHz dynamic frequency divider that were successfully fabricated using the 148-GHz fmax technologies are described.

We report the development of high-performance small-scale AlGaAs/GaAs collector-up heterojunction bipolar transistors (C-up HBT) with a carbon (C)-doped base layer. Oxygen-ion (O+) implantation is used to define their intrinsic emitter/base junctions and zinc (Zn)-diffusion is used to lower the resistivity of their O+-implanted extrinsic base layers. The highly resistive O+-implanted AlGaAs layer in the extrinsic emitter region sufficiently suppresses electron injection even under high-forward-bias conditions, allowing high collector current densities. The use of a C-doped base is especially effective for small-scale C-up HBT's because it suppresses the undesirable turn-on voltage shift caused by base dopant diffusion in the intrinsic area around the collector-mesa perimeter that occurs during the high-temperature Zn-diffusion process after implantation. Even in a small-scale trasistor with a 2 µm2 µm collector, a current gain of 15 is obtained. A microwave transistor with a 2 µm10 µm collector has a cutoff frequency fT of 68 GHz and a maximum oscillation frequency fmax of 102 GHz. A small-scale C-up HBT with a 2 µm2 µm collector shows a higher fmax of 110 GHz due to reduced base/collector capacitance CBC and its fmax remains above 100 GHz, even at a low collector current of 1 mA. The CBC of this device is estimated to be as low as 2.2 fF. Current gain dependence on collector size is also investigated for C-up HBT's and it is found that the base recombination current around the collector-mesa perimeter reduces the current gain.

Self-aligned InP/InGaAs heterojunction bipolar transistors (HBTs) were fabricated with emitter electrodes of 12, 22, 25, and 220 µm2 on the same wafer to investigate the influence of lateral scaling on device performance. DC characterization of these devices showed that InP/InGaAs HBTs are less subject to the emitter-size effect than GaAs-based HBTs. Common-emitter current gain β of the smallest 12-µm2 transistor was approximately 60 which is high enough for practical use. High-frequency characteristics of the transistors were almost the same in spite of the large difference in device size. Unity current-gain cutoff frequency fT of the smallest 12-µm2 transistor was as high as 163 GHz at a collector current of 2.3 mA, which ranks with the fT176 GHz achieved by the largest 220-µm2 transistor at a collector current of 45 mA. The smallest device also showed an excellent high-speed performance of fT100 GHz at submilliampere collector currents of Ic0.6 mA. The results indicate that small-lateral-dimension InP/InGaAs HBTs are applicable to high-speed ICs with low power dissipation.

Measuring the relational similarity between word pairs is important in numerous natural language processing tasks such as solving word analogy questions, classifying noun-modifier relations and disambiguating word senses. We propose a supervised classification method to measure the similarity between semantic relations that exist between words in two word pairs. First, each pair of words is represented by a vector of automatically extracted lexical patterns. Then a binary Support Vector Machine is trained to recognize word pairs with similar semantic relations to a given word pair. To train and evaluate the proposed method, we use a benchmark dataset that contains 374 SAT multiple-choice word-analogy questions. To represent the relations that exist between two word pairs, we experiment with 11 different feature functions, including both symmetric and asymmetric feature functions. Our experimental results show that the proposed method outperforms several previously proposed relational similarity measures on this benchmark dataset, achieving an SAT score of 46.9.

Device figure-of-merits for digital ICs are derived from analytical delay expressions for emitter-coupled logic and source-coupled FET logic inverters and are compared with the operating speeds of D-F/Fs reported in previous studies. We show that device figure-of-merits for baseband amplifiers are equivalent to those for digital ICs. The validity of device figure-of-merits are confirmed by measuring the bandwidth of the baseband amplifiers fabricated with AlGaAs/GaAs LBCTs.

We fabricated a 2:1 multiplexer IC (MUX) with a retiming function by using 1-µm self-aligned InP/InGaAs/InP double-heterojunction bipolar transistors (DHBTs) with emitter mesa passivation ledges. The MUX operated at 120 Gbit/s with a power dissipation of 1.27 W and output amplitude of 520 mV when measured on the wafer. When assembled in a module using V-connectors, the MUX operated at 113 Gbit/s with a 514-mV output amplitude and a power dissipation of 1.4 W.

A compact optically-fed radio access point module was developed that consists of a uni-traveling-carrier refracting-facet photodiode, a patch antenna, and an optical input interface. An output power from the photodiode was 1.4 dBm at a frequency of 5.88 GHz without any bias voltage.

We fabricated low-jitter 2:1 multiplexer (MUX) and 1:2 demultiplexer (DEMUX) modules for bit error rate testers that can be used for research into ultra-high-bitrate communication subsystems and devices with bitrates of over 100 Gbit/s. The 1:2 DEMUX IC design took into consideration an IC layout allowing module pin placement for optimal utility. With regard to mounting, the 2:1 MUX and 1:2 DEMUX modules were constructed using transmission lines of grounded coplanar waveguide (G-CPW) configuration, which offers excellent high-frequency characteristics. These modules operated at 113 Gbit/s with a low root mean square jitter of 548 fs and 587 fs, respectively.