Explore the vast range of titles available.

One-class collaborative filtering (OCCF) is an emerging setup in collaborative filtering in which only positive examples or implicit feedback can be observed. Compared with the traditional collaborative filtering setting where the data have ratings, OCCF is more realistic in many scenarios when no ratings are available. In this paper, we propose to improve OCCF accuracy by exploiting the rich user information that is often naturally available in community-based interactive information systems, including a user’s search query history, and purchasing and browsing activities. We propose two major strategies to incorporate such user information into the OCCF models: One is to linearly combine scores from different sources, and the other is to embed user information into collaborative filtering. Furthermore, we employ the MapReduce framework for similarity computation over millions of users and items. Experimental results on two large-scale retail datasets from a major e-commerce company show that the proposed methods are effective and can improve the performance of the OCCF over baseline methods through leveraging rich user information.

With the advancement of deep cryogenic detection technology, spacecraft are required to operate at a background temperature of 100 K or lower, necessitating the use of pulse tube cryocooler as a critical support component. Traditionally, the compressor and hot end heat exchanger of pulse tube cryocooler function at an ambient temperature of 300 K. Multi-stage pulse tube cryocoolers typically require precooling to a temperature range of 80 K to 100 K before the second stage can commence operation. The transient regenerator serves as the thermal buffer between the ambient temperature compressor and the secondary pulse, leading to significant PV power losses and reduced cryocooler efficiency. Additionally, two-stage pulse tube cryocoolers often exhibit low operating frequencies, large volumes and weights, and high launch costs. This paper presents the design of a micro single-stage coaxial pulse tube cryocooler capable of direct operation in the 80 K temperature range. The cryocooler is powered by liquid nitrogen precooling and a linear compressor, with a total mass of 2.6 kg. It employs inertance tube and gas reservoir as phase shifters. The cold finger has a diameter of 14 mm and a fill length of 55 mm. Preliminary experiments yielded the following results: at an operating frequency of 60 Hz, an input power of 20 W, a hot end temperature of 80 K, and an operating pressure of 1.5 MPa, the minimum no-load temperature achieved was 13 K, and a cooling capacity of 1 W at 28.2 K was obtained at 52 Hz.

Lightweight pulse tube cryocooler holds significant potential for cooling infrared detectors in various applications. Previous research on Lightweight pulse tube cryocoolers has been reviewed in this work, and a theoretical analysis of key parameters for achieving lightweighting has been conducted. Meanwhile, experimental investigations were carried out to optimize the phase shifter (inertance tube) and charging pressure. Various combinations of inertance tubes were tested, and the charging pressure was systematically varied. The optimized cryocooler can obtain 2.9W of cooling capacity at 80K under the input power of 60W and an optimal frequency of 104 Hz, with a relative Carnot efficiency of 13.07% and a weight of only 1.6kg, and the specific mass reached 1.81 W/kg, which is at a high level among the 80K miniature pulse tube cryocoolers reported so far.

The Stirling type pulse tube cryocooler (SPTC) eliminates moving parts at the cold end and is driven by a linear compressor at the hot end, thus offering the advantages of low vibration at both ends, high reliability, and long service life, which makes it attractive for various special fields, such as the space field.In practice, there is an increasing interest in providing cooling power at different temperature levels. This paper presents an experimental investigation of a thermal-coupled two-stage pulse tube cryocooler with multi-bypass structures. The experiments investigated the effects of the second-stage operating frequency, charging pressure, and input power on performance. Building upon this, the study explored the interaction between the first-stage temperature, multi-bypass temperature, and second-stage temperature, alongside experimental testing of cooling performance. The experimental results show that through the thermal coupling pre-cooling and the design of multi-bypass structure, the second-stage cold finger can realize the lowest temperature of 5.2 K and the cryocooler can provide 54 mW cooling capacity at 8 K while providing 50 mW cooling capacity at 35 K with the 1st-stage providing 1.5 W of cooling capacity.

The operating temperature of 30 K is crucial for long-wave infrared and very long-wave infrared detectors. To achieve cooling at 30 K, a two-stage pulse tube cryocooler (PTC) has been employed in most previous studies. However, these systems are complex in structure. Single-stage 30 K high-efEiciency PTC developments present signiEicant challenges but are essential for 30 K long-wave infrared detectors. We designed a single-stage coaxial PTC with a double inlet oriEice and conducted extensive optimization experiments on this cryocooler. With 200 W input power and a hot end temperature of 285 K, the PTC achieved a cooling capacity of 1.6 W/30 K, corresponding to a relative Carnot efEiciency of 6.8%.

To develop a long-wavelength infrared (LWIR) detector, a 40 K pulse tube cryocooler is needed to provide reliable and low-noise cooling power. Traditionally, it is generally believed that to improve efficiency, a 40 K pulse tube cryocooler needs to operate at around 40 Hz. However, low frequency cryocoolers are heavier and are not preferred for use on satellites. In this paper, a lightweight pulse tube cryocooler working at 76 Hz is designed. The inertance tubes, along with the reservoir, serve as the only phase-shifter to guarantee the stability. Currently, the cryocooler has a cooling capacity of 9 W at 40 K while operating at 500 W of electrical power, and it weighs only 7.8 kg. The efficiency relative to the Carnot efficiency was approximately 11.1%. The performance characteristics of the designed cryocooler are presented in detail.

The regenerator is a crucial component in regenerative cryocoolers, responsible for providing cooling power through the heat exchange between helium and the regenerative materials within it. When the temperature drops below 20 K, the specific heat capacity of helium increases and surpasses that of the stainless steel, impeding the heat transfer process within the regenerator and limiting the enhancement of cooling power in cryocoolers operating at 20 K. To enhance the cooling power of a two-stage thermally-coupled pulse tube cryocooler functioning within the 20 K temperature range, this paper focuses on optimizing the regenerative materials. Stainless steel screens, Er-plated stainless-steel screens, and pressed Er-plated stainless-steel screens were used as regenerative materials. The parameters and experimental results of these materials were compared.

With the rapid advancement of space technology, there is a continuous increase in demand for long-wave infrared detection devices, leading to higher requirements for two-stage pulse tube cryocoolers operating at liquid hydrogen temperatures. The performance of the cryocoolers is closely related to the inertance tube size. In order to improve a two-stage thermally-coupled pulse tube cryocooler’s performance in the liquid hydrogen temperature range, this study conducted phase optimization experiments, analyzed the effects of different inertance tube combinations on the cryocooler’s performance. The optimal operation frequency, power factor, no-load cooling temperature, cooling capacity and rCOP were compared for cryocoolers with different inertance tube combinations. After preliminary optimization, the cryocooler achieved 586 mW/20 K cooling power with a total electric power consumption of 185 W, resulting in a rCOP of 4.43%.

Search and information retrieval technologies have significantly transformed the way people seek information and acquire knowledge from the internet. To further improve the search accuracy and usability of the current-generation search engines, one of the most important research challenges is for a search engine to accurately understand a user’s intent or information need underlying the query. This thesis presents a systematic study of query understanding. In this thesis I have proposed a conceptual framework where there are different levels of query understanding. And these levels of query understanding have natural logical dependency. After that, I will present my studies on addressing important research questions in this framework. First, as a major type of query alteration, I addressed the query spelling correction problem by modelling all major types of spelling errors with a generalized Hidden Markov Model. Second, query segmentation is the most important type of query linguistic signals. I proposed a probabilistic model to identify the query segmentations using clickthrough data. Third, synonym finding is an important challenge for semantic annotation of queries. I proposed a compact clustering framework to mine entity attribute synonyms for a set of inputs jointly with multiple information sources. And finally, in the dynamic query understanding, I introduced the horizontal skipping bias which is unique to the query auto-completion process (QAC). I then proposed a novel two-dimensional click model for modeling the QAC process with emphasis on such behavior.

Background Home visiting nurses (HVNs) have long been part of home and community-based care interventions designed to meet the needs of functionally declining older adults. However, only one of the studies including HVNs that have demonstrated successful impacts on Activities of Daily Living (ADL) has reported how those interventions affected individual ADLs such as bathing, instead reporting the effect on means of various ADL indices and scales. Reporting impacts on means is insufficient since the same mean can consist of many different combinations of individual ADL impairments. The purpose of our study was to identify which individual ADLs were affected by a specific HVN intervention. Methods This is a secondary analysis comparing two arms of a randomized controlled study that enrolled Medicare patients (mean age = 76.8 years; 70% female) with considerable ADL impairment. At baseline difficulty with individual ADLs ranged from a low of 16.0% with eating to a high of 78.0% with walking. Through monthly home visits, the HVN focused on empowering patients and using behavior change approaches to facilitate chronic disease self-management. Three categories of analyses were used to compare difficulty with and dependence in 6 individual ADLs between the HVN (n = 237) and care as usual (n = 262) groups (total N = 499) at 22 months after study entry: (1) unadjusted analyses that strictly depend on random assignment, (2) multinomial logistic regression analyses adjusting for baseline risk factors, and (3) multinomial regression analyses that include variables reporting post-randomization healthcare use as well as the baseline risk factors. Results Compared to care as usual, patients receiving the HVN intervention had less difficulty performing bathing at 22 months. However, there were no effects for difficulty performing the other 5 ADLs. While no effects were found for lower levels of dependence for any ADLs, impacts were detected for the most dependent levels of 4 ADLs: patients experienced less dependence in walking and transferring, a substitution effect for toileting, and more dependence in eating. Conclusions Future research is needed to confirm these findings and determine how HVN interventions affect individual ADLs of older adults with multiple ADLs.