Documentation Matters: Human-Centered AI System to Assist Data Science Code Documentation in Computational Notebooks

Computational notebooks allow data scientists to express their ideas through a combination of code and documentation. However, data scientists often pay attention only to the code, and neglect creating or updating their documentation during quick iterations. Inspired by human documentation practices learned from 80 highly-voted Kaggle notebooks, we design and implement Themisto, an automated documentation generation system to explore how human-centered AI systems can support human data scientists in the machine learning code documentation scenario. Themisto facilitates the creation of documentation via three approaches: a deep-learning-based approach to generate documentation for source code, a query-based approach to retrieve online API documentation for source code, and a user prompt approach to nudge users to write documentation. We evaluated Themisto in a within-subjects experiment with 24 data science practitioners, and found that automated documentation generation techniques reduced the time for writing documentation, reminded participants to document code they would have ignored, and improved participants’ satisfaction with their computational notebook.

Abstract 355: Using Low-Fidelity Simulation with Hospital Specific Equipment to Improve Front Line Nursing Staff Knowledge, Skill, and Confidence in a Code Blue

Introduction: At this facility, staff response to unresponsive victims prior to arrival of Rapid Response Team is unreliable. The gaps in care reported by debrief forms include unable to find and use equipment on the crash cart, unable to appropriately document Code Blue in EHR, and not performing adequate CPR. Currently, the nursing staff is required to maintain Basic Life Support (BLS) and Advanced Life Support (ALS) (per their job description) via the Resuscitation Quality Improvement (RQI) program. RQI delivers quarterly training via computer scenarios and mannequin simulations. There is no consideration in RQI for hospital specific equipment or processes. The recommendation for improving staff response to code situations is to incorporate an annual Mock Code Event for all nursing staff to improve performance during code situations, in addition to RQI quarterly activities. Hypothesis: Participating in an instructor led Mock Code Event annually to review hospital specific equipment and processes, along with the quarterly RQI simulations, will improve staff response to code blue situations. Methods: The Mock Code Event included stations to review items in crash cart, use of defibrillator, documenting in EHR, and review of BLS. After completing the stations, the staff participated in a low-fidelity code blue simulation with a debrief. At end of activity, participants voluntarily completed a survey regarding Mock Code Event. Results: The participants scored 6 items on a Likert scale (from 1 disagree to 5 agree). 457 RNs and PCTs attended Mock Code Events in 2019. 189 completed survey after event. Conclusion: The survey findings demonstrate the Mock Code Event was an effective method to improve knowledge, skill, and confidence of front-line staff during Code Blue situations. The survey findings support that the educational techniques used were helpful and easy to understand.

A New Estimation Method of Transient Voltage Stability Limit Clearing Time

Critical clearing time index is one of the indices to assess transient voltage stability. At present it is mainly calculated based on the instantaneous value after transient voltage instability or the steady state valueafer fault clearing time. So a new method to estimate the critical clearing time is presented which is correspongding to the induction motor minimum slipvalue in the fault. It can forecasttransient voltage instability,and has higher precision comparing whithquik criterion and analytical method. The simulation results of IEEE 30-bus system verify the effectiveness of the stated method. .CLC number: TM 73 Document code: A Article ID: 000-3673(2013)00-0000-00

Influence of Moisture Content to the Freeze-Thaw Performance of Roadbed in High-Cold Areas

Frost-heave and thaw-settlement of roadbed soil in highway will influence directly the durability, safe traffic flow and construction & maintenance costs in high-cold areas, therefore, recognizing and analysing the common embankment technologies of highway roadbed in high-cold areas accurately is significant to the effective controlling of project invest and the highway construction with limited funds in minority areas. The relations of Moisture Content and the freeze-thaw performances of roadbed fillers, subgrade soil were got respectively by experiments, and the results shows: Moisture Content has larger influence on the frost-heave and thaw-settlement performance of the soil. During the embankment of roadbed, the Moisture Content of fillers should be controlled nearby the optimum Moisture Content. The frost-heave and thaw-settlement occurs mainly in the subgrade soil, controlling the Moisture Content of subgrade soil is very important to improve the up-limit of frozen-soil, keep the stability of frozen-soil, control the thaw-settlement of roadbed and get rid of the roadbed diseases. CLC: U416.1 Document code: B

A Knowledge-Based Approach of Automation Design of Aircraft Standard Components

To solve the problem that knowledge of standard components cannot be quickly applied to design process of aircraft, a novel approach based on knowledge is proposed from the instantiated requirements of standard components. Knowledge-based model based on semantic of standard component is described, which is composed of parameters, rules, solid model and assembly information. The information model is expressed and stored based on XML. Knowledge-based model is provided to support the specifications determination, semantic modeling of solid model and batch assembly of standard components by loosely coupled manner. Design knowledge is seamlessly, quickly applied to the instantiated process of standard components. Effectiveness of the proposed model and method is verified by an aircraft design example of standard components. CLC number: TP391.7 Document code: A

Research on Searching Guidance of TV Guided Missile

The searching phase of TV guided missile is usually divided into three different stages as the navigation, homing and terminal guidance. In this paper, guidance laws in different phases are analyzed. To solve the problem of target acquisition, signal following and navigation precision during the period of navigation, methods of integrated guidance with three geostationary satellites orientation and INS in high dynamic circumstances are put forward. The mathematics models of pre - guiding and searching are established corresponding to the traits of smaller TV seeker instantaneous field, limited shooter reacting velocity, low hunting speed and narrow searching area. The compound guidance methods which integrate the tracking and the proportional in terminal phase are raised in order to increase the guiding precision and to decrease the undershoot value. Experiment shows that the guiding precision is raised by a long way.CLC number: TJ765.4+3 Document Code: B

Experimental Study of Finding Best Way of Radiation Seeds with HVEF

Norder to find the best high-voltage electrostatic radiation seed program, to select two different kinds of cucumbers, vegetables, paddy rice seeds, each seed be divided into 5 experimental groups, and be treated by high voltage electrostatic field (HVEF) at different field strength and time, and testing and statistics of the seeds vigour at germination stage. The results indicated that the greater the radiation field to same seed, the shorter radiation time is required, and vice versa. For each species, E•t tends to a constant (the constant is different in different species). Be proved by mathematical derivation that the HVEF radiation transport essentially energy to the seed, and then will be able to find the best way for HVEF radiation seeds. CLC number: S129 Document code: A.