Experimentation Sample Clauses

Experimentation. The Board and the Association recognize the need for experimentation and innovation in instructional programs and techniques and agree to cooperate in the implementation thereof.

Experimentation. Adaptive management is strongly rooted in scientific experimentation. By specifically designing experiments into management actions, conclusions can be drawn that help develop better resource management decision making. Experimentation in Battle Creek is embodied in three ways, where experimentation (1) has been a component of adaptive management problem definition and solution development, (2) is embodied in the overall Adaptive Management program as envisioned in this document, and (3) may be conducted as part of individual Adaptive Management objectives considered under this plan within the established protocols.

Experimentation. In order to accurately understand the role of hormone status and relative levels of progesterone-derived neurosteroids, it was important to have the mice on a normal estrous cycle. This cycle occurs on a shorter time scale than the 28- day human menstrual cycle. A typical cycle will last 4 days and pass through 4 stages, with diestrus experiencing sharp changes in hormone levels and estrus showing lower, more stable levels (▇▇▇▇▇▇▇ et al., 1974). For the most part, when female mice are included in an experiment, their hormonal levels do not accurately mimic what they would be in nature. There is a phenomenon called the ▇▇▇-Boot Effect, where females that are in same-sex housing will have suppressed estrous cycles (▇▇▇ ▇▇▇ ▇▇▇, 1955). The ▇▇▇▇▇▇▇ Effect has to do with the initiation of ovulation and the regular cycling of ovarian hormones by the presence of male pheromones (▇▇▇▇▇▇▇ et al., 1957). For this project, nest materials were taken out of a dirty male cage and placed in a cage that housed a group of females. Ovulation typically happens about a day after the cage is spiked with the male bedding. The females were allowed to naturally cycle for a week before testing began. Estrus females had their bedding spiked about 5-6 days prior to testing, whereas diestrus females had their bedding spiked about 7-8 days prior to testing. On the morning that testing was to take place, vaginal lavage samples were taken by flushing a small amount of saline into the vaginal opening to collect a sample of epithelial cells. The slides were examined to ensure the status of the females, before being allowed to dry prior to staining. Due to the stressful nature of the lavage samples, the males were also individually handled to equate the experiences of the two sexes. After the slides had dried completely, a vaginal cell staining protocol was followed from ScyTek (Papanicolaou Staining Protocol). The smears were then imaged and classified, based on cell cytology information and correlating hormone levels (▇▇▇▇▇▇ et al., 2012). The females were then divided into two groups based on cell composition of the vaginal lavage samples and the correlating levels of hormones. Estrus females were classified if the majority of the cells in the vaginal lavage sample were cornified epithelial cells, stained pink; diestrus females were categorized by a predominance of leukocytes, stained dark brown. The first experiment that was carried out was to look at the anxiolytic effect of either o...

Experimentation. 5.1 Case a1: Adaptive traffic sampling and management 5.1.1 Performance objectives, and evaluation criteria Collected measurements can decide on how to tune the sampling rates inside the network. We consider for this purpose an important monitoring application, the estimation of the volume (in terms of number of packets or bytes) of some chosen network flows. A flow is a set of 5-tuple flows that share some common features as the source IP address prefix, the destination IP address prefix, the same ingress and egress routers inside the network, etc. At the limit, a flow can be one 5-tuple flow or even the entire network traffic. Given a set of flows to monitor, the machine learning engine should progressively tune the sampling rates in routers in such a way to minimize the global estimation error.

Experimentation. Section 1. The Board and the Federation recognize that a sound educational program requires not only the efficient use of existing resources but also constant experimentation with new methods and organization. The Federation agrees that experimentation presupposes flexibility in assigning and programming pedagogical and other professional personnel. Hence, the Federation will facilitate its members' voluntary participating in new ventures that may depart from usual procedures. The Board agrees that educational experimentation will be consistent with the standards of working conditions prescribed in this agreement. Section 2. All proposals for educational experimentation which require modification of class size limits shall be studied by a committee comprised of two (2) persons selected by the Superintendent and two (2) persons chosen by the Federation and a third teacher chosen by die other four. Section 3. All proposals under Section 2 shall contain: a) Educational objectives of the experiment b) Scope (Number of students and teachers involved) c) Evaluation criteria and procedures

Experimentation. Expedient provides a useful graphical interface to the users for managing their experiments. Through Expedient, a user can create a project (a container for experiments) by sending a request to the administrator. Once the project is created, the creator can add other users to the project so that they can also perform experiments in that project. The user can also add aggregate managers or RMs to the project, that is, the resources that will be available for the experiments. The experiments are created as slices. AGer creating a slice, a user can select from AMs provided by that project for use in experiment. Once included, the resources of these AMs are graphically displayed in Expedient along with the links between them. A new GUI section for the Expedient has been created in ▇▇▇▇▇ project which provides information about the status of the components and the slice topology.

Experimentation. To demonstrate the effectiveness of cause and effect phrase extraction when using Causal Cue Phrases, we propose to compare accuracy of two Conditional Random Fields Relations Learning Algorithms (▇▇▇▇▇▇ and ▇▇▇▇▇▇▇▇, 2006) trained to perform a token level labeling of cause and effect phrases. One of these CRFs will be trained given access to CCP knowledge in the form of a single additional Boolean feature. In all other respects these two algorithms will be identical. Because it is only possible to extract correct cause/effect phrases from sentences containing a causal relation, the training set for CRF will consist of all causal sentences in the corpus. Training and testing for CRF will take place using a ten-fold cross validation method. Both algorithms will perform extraction on a token level, classifying each token as one of three distinct classes: cause (part of cause phrase), effect (part of effect phrase), and non-causal (part of neither cause nor effect phrase). To evaluate the performance of each algorithm, a phrased based accuracy metric will be employed. An algorithm will be said to have correctly identified the cause phrase of a sentence if the cause phrase it identified overlaps with expected cause phrase but does not overlap with the expected effect phrase. Similarly, to correctly identify the effect phrase, it is necessary to overlap with the expected effect phrase but not the expected cause phrase. Additionally, we employ a token based accuracy metric to provide a stricter evaluation. Each token is said to be correct if matches with the expected class label. One drawback of using such a token based accuracy metric is the ability to get rather high accuracy scores simply by marking every token as Non-Causal. To combat this, we introduce a third accuracy metric, termed Focused Token Accuracy which resembles the usual token based accuracy metric only it is computed only over the accepted cause and effect phrases. This further step eliminates all Non- Causal tokens, and reveals how well each algorithm is able to perform attempting to label only cause and effect phrases. The CRF will be trained with the following feature set: Part of Speech: the part of speech of the current token Stem: the stem of the current token VP or NP: whether the current phrase (most direct parent) is a noun phrase or verb phrase Adjacent word information: the words and features of 3 adjacent tokens For the CRF trained with CCP knowledge, we add a single additional featu...

Experimentation. In the case of long-term experimentation, the Union shall receive prior notice thereof.

Experimentation. OxyChem has committed to perform a study of fugitive mercury emissions from the mercury cell building within its chlor-alkali facility. OxyChem will accomplish this study by installing a MMS Mercury Monitoring System supplied by Mercury Instruments, GmbH, of Germany. This instrument will draw samples from sixteen monitoring points. The system utilizes an Ultra Violet Photometer to detect mercury based on the UV absorption of mercury at 253.7 nm. Pilot studies conducted by OxyChem utilizing this technology have produced favorable results. Four (4) sample points will be monitored in each of the four quadrants of the cell building to provide a baseline for establishing average mercury concentrations that will be correlated with representative air flow calculations for the cell building. The monitored concentrations will be integrated into the facility’s realtime DCS that calculates air flow through the building. This is expected to result in the ability to estimate the mass of fugitive mercury emissions emitted from the cell building via the roof ridge vent system. The data from the individual monitoring points will be analyzed to localize instances of higher mercury concentrations. This localization will provide the ability to analyze what activities may be causing increased emissions of fugitive mercury. These activities can then be studied to reduce these emissions.