scholarly journals Re-ranking the Twitter search result.

Author(s):  
Prerna Pathak

Social Sharing Platforms, a great source of free and diverse information, have been center of attraction to many people. Users post their opinions, thoughts, life events, news and all other information. This data flowed into these systems has increased to such a limit making nearly impossible for a user to read all or even most of it, analyzing and utilize it. As a solution to this problem, we here are proposing an approach, which makes use of not only the tweets themselves but also their properties to re-rank the tweets given a user query. The proposed approach was implemented in a prototype system and test results were generated. A set of feedback data collected via online survey for those test results provides a good evaluation score, with an average improvement of around 10% on precision values after removing the outliers. It shows that our approach can generate improved results over the original ones.

2021 ◽  
Author(s):  
Prerna Pathak

Social Sharing Platforms, a great source of free and diverse information, have been center of attraction to many people. Users post their opinions, thoughts, life events, news and all other information. This data flowed into these systems has increased to such a limit making nearly impossible for a user to read all or even most of it, analyzing and utilize it. As a solution to this problem, we here are proposing an approach, which makes use of not only the tweets themselves but also their properties to re-rank the tweets given a user query. The proposed approach was implemented in a prototype system and test results were generated. A set of feedback data collected via online survey for those test results provides a good evaluation score, with an average improvement of around 10% on precision values after removing the outliers. It shows that our approach can generate improved results over the original ones.


Author(s):  
James F. Walton ◽  
Andrew Hunsberger ◽  
Hooshang Heshmat

In this paper the authors will present the design and preliminary test results for a distributed electric generating system that uses renewable energy source for economical load-following and peak-shaving capability in an oil-free, high-speed micro-turboalternator system using compliant foil bearings and a permanent magnet alternator. Test results achieved with the prototype system operating to full speed and under power generating mode will be presented. A comparison between predicted and measured electrical output will also be presented up to a power generating level of 25 kWe at approximately 55,000 rpm. The excellent correlation between design and test provides the basis for scale up to larger power levels. Based upon the turboalternator test results a thermodynamic cycle analysis of a system using low grade waste heat water at approximately 100 C will be reviewed. The tradeoff study results for a series of environmentally friendly refrigerant working fluids will also be presented including sensitivity to vaporization and condensing temperatures. Based on the cycle and pinch point analyses predicted maximum output power was determined. Finally a preliminary turbine design for the selected R134a working fluid was completed. The results of this study show that a net output power level of greater than 40 kW is possible for approximately 240 l/m flow of water at 100C is possible.


2019 ◽  
Vol 21 (2) ◽  
pp. 70-75
Author(s):  
Marnina Stimmel ◽  
Shaina Shagalow ◽  
Elizabeth K. Seng ◽  
Jeffrey G. Portnoy ◽  
Roseann Archetti ◽  
...  

Abstract Background: Adherence to nonmedication recommendations is typically low, as seen in various health populations. Because literature on adherence to treatment recommendations made after neuropsychological testing in multiple sclerosis (MS) is lacking, this study evaluated adherence and reasons for nonadherence. Relationships between adherence to recommendations and various other factors in patients with MS were also evaluated. Methods: Of 66 adult patients seen for neuropsychological testing at an MS center, 55 were eligible for this study. Forty-five patients (mean age, 43.4 years; 75.6% women) were reached by phone, and all agreed to an interview involving questions regarding adherence to treatment recommendations. Other information was obtained through retrospective medical record review. Results: Overall self-reported adherence to recommendations made from neuropsychological testing was 38%. Adherence rates varied by recommendation type: psychopharmacological management had the highest rate (80%), and referrals for cognitive rehabilitation had the lowest (6.5%). Reasons for nonadherence included needing more information and wanting to speak with one's physician regarding the recommendations. Adherence was associated with patients' ability to spontaneously recall at least some of their recommendations and with receiving both a written report and a phone call with test results. Conclusions: Adherence to recommendations made after neuropsychological testing for patients with MS is low. Points of intervention may be to give directed feedback for each recommendation and to provide both a written report and a phone call with results and recommendations. Asking patients to repeat back the recommendations may be a simple and efficient way to increase understanding and improve adherence.


2015 ◽  
Vol 1092-1093 ◽  
pp. 175-180
Author(s):  
Dong Lai Xie ◽  
Bing Qi Wang

Fuel cell based micro combined heat and power (micro-CHP) systems are residential scale clean energy conversion unit. It employs fuel cells in a compact system that converts natural gas, propane or other fuels into both electricity and heat, which increases efficiency by simultaneously generating power and heat for one unit, on-site within a home. A prototype system consisting of a natural gas steam reforming unit, CO cleaning unit, PEM fuel cell stack, waste heat recovery unit and auxiliary unit is integrated. Test results of the prototype show that it can start within an hour and the syngas produced can meet the fuel cell’s requirements. The prototype’s electric power and thermal power are 200W and 530W respectively, while the electric and thermal efficiency are 15.4% and 40.9% respectively.


2019 ◽  
Vol 29 (2) ◽  
pp. 334-345 ◽  
Author(s):  
Janne Cadamuro ◽  
Michael Cornes ◽  
Ana-Maria Simundic ◽  
Barbara de la Salle ◽  
Gunn B.B. Kristensen ◽  
...  

Introduction: No guideline currently exists on how to detect or document haemolysis, icterus or lipemia (HIL) in blood samples, nor on subsequent use of this information. The EFLM WG-PRE has performed a survey for assessing current practices of European laboratories in HIL monitoring. This second part of two coherent articles is focused on HIL. Materials and methods: An online survey, containing 39 questions on preanalytical issues, was disseminated among EFLM member countries. Seventeen questions exclusively focused on assessment, management and follow-up actions of HIL in routine blood samples. Results: Overall, 1405 valid responses from 37 countries were received. A total of 1160 (86%) of all responders stating to analyse blood samples - monitored HIL. HIL was mostly checked in clinical chemistry samples and less frequently in those received for coagulation, therapeutic drug monitoring and serology/infectious disease testing. HIL detection by automatic HIL indices or visual inspection, along with haemolysis cut-offs definition, varied widely among responders. A quarter of responders performing automated HIL checks used internal quality controls. In haemolytic/icteric/lipemic samples, most responders (70%) only rejected HIL-sensitive parameters, whilst about 20% released all test results with general comments. Other responders did not analysed but rejected the entire sample, while some released all tests, without comments. Overall, 26% responders who monitored HIL were using this information for monitoring phlebotomy or sample transport quality. Conclusion: Strategies for monitoring and treating haemolytic, icteric or lipemic samples are quite heterogeneous in Europe. The WG-PRE will use these insights for developing and providing recommendations aimed at harmonizing strategies across Europe.


1970 ◽  
Vol 111 (5) ◽  
pp. 15-18
Author(s):  
A. Kajackas ◽  
V. Batkauskas ◽  
A. Saltis ◽  
D. Gursnys

A project of creation of a multi-dimension and autonomous QoS evaluation and monitoring system is presented in the paper. The initial data for evaluation of the QoS is collected from users using specially created software programs - iQoS modules. Singularities of the initial data collection are discussed. Test results of the prototype system are presented. Ill. 2, bibl. 14 (in English; abstracts in English and Lithuanian).http://dx.doi.org/10.5755/j01.eee.111.5.347


Author(s):  
S. OHARA ◽  
F. TSUNODA ◽  
H. MAEZAWA ◽  
A. K. ONOMA ◽  
M. HUI ◽  
...  

To assure the quality of software by running test cases and evaluating the results is one of the difficult parts of the entire software development project. The difficulty usually comes from the lack of appropriate supporting tools and the complexity of the software. In the past ad hoc supporting tools were made for each project and test results were usually not used across projects. This conventional way of test and evaluation (T&E) is time consuming, and the most important decision "When is this software ready to ship?" is left to the engineers depending on their experiences. Our objective is to build a knowledge-based T&E environment such that tests cases, test results, object snapshots and other information are accumulated in a database. These longitudinal data can be automatically tracked and analyzed to provide decision support information. As a result, test results can be reviewed repeatedly and software quality can be assured by analyzing these data from various perspectives.


Author(s):  
BO-REN BAI ◽  
BERLIN CHEN ◽  
HSIN-MIN WANG

In light of the rapid growth of Chinese information resources on the Internet, this study investigates a novel approach that deals with the problem of Chinese text and spoken document retrieval using both text and speech queries. By properly utilizing the monosyllabic structure of the Chinese language, the proposed approach estimates the statistical similarity between the text/speech queries and the text/spoken documents at the phonetic level using the syllable-based statistical information. The investigation successfully implemented a prototype system with an interface supporting some user-friendly functions and the initial test results demonstrate the feasibility of the proposed approach.


2017 ◽  
Vol 5 (2) ◽  
pp. 59-64
Author(s):  
Annisau Saidah

Storage of wheat flour requires good temperature and humidity to avoid mold growth that can damage the nutrients in wheat flour. The temperatures and humidity required for proper storage are above temperatures and humidity below 60% RH. The method used in flour storage is FIFO (First In First Out) using a Load cell sensor as a marker for flour input. The incoming flour will input the time, location and amount of flour, so that it can be easily determined which flour will come out first. To facilitate monitoring of temperature, humidity and stock amount in flour storage in this study, sending monitoring data wirelessly using an internet network that can be accessed anywhere. The test results of this study show that the DHT22 sensor can measure temperature and humidity well, which has a RE (Relative Error) value of 0.5% temperature and 3.9% humidity compared to the HTC-1 measuring instrument. The load cell test compared to digital scales has an accurate value using a linear formula. In the QoS test, sending sensor data to the web server has a good value according to the ITU-T standard, namely delay <150ms


Sign in / Sign up

Export Citation Format

Share Document