The term functional grammar has been used before, notably by Dik (1978). I risk adding to the number of its meanings here, and thus debasing its value, only because it is peculiarly apt for this new employment. I propose to outline a new grammatical formalism which, if it can be successfully developed, will be worthy of the name functional on three counts. First, it is required to function as part of a model of language production and comprehension. The formalism is interpretable by an abstract machine whose operation is intended to model tl1c syntactic processing of sentences by speakers and hearers indifferently. This is not to say that it is not also intended to represent a speaker's grammatical competence. Secondly, the formalism ascribes to every sentence, word, and phrase, a functional description which differs from the strnctural description of better known formalisms mainly by stressing the function mat a part plays in a whole rather than me position a part occupies in a sequence of omers. The names of grammatical categories, like S, NP, and VP will therefore play a secondary role to terms like subject, object, and modifier. Thirdly, properties that distinguish among logically equivalent sentences will have equal importance with properties that they share. In omer words, mis will be a functionalist view of grammar in which notions like topic and focus, given and new will have equal status with subject and predicate, positive and negative.For me most part, theoretical linguists see a grammar as an abstract device mat characterizes the presumably infinite set of sentences of a language, mat is, which differentiates tl1c sentences from omer strings which are not sentences. Computational linguists, on me otl1cr hand, have usually taken a grammar to be a transducer showing how a meaning comes to be represented as a string of words or, more frequently, how a string of words is analyzed to reveal its meaning. Functional grammar has both aspects. It can also be said to be a transducer whose input is a more or less incomplete account of me syntactic relations among me parts of a sentence and whose output is one or more accounts which are complete according to me meory. Given a more or less incomplete description, it verifies mat it describes a legal grammatical object-a word, phrase, or sentence-and adds such additional detail as me grammar allows. If it is not a legal grammatical object, no output is produced. If it is, one or more descriptions are produced, each an enrichment of the original, but reflecting different grammatical interpretations.The ideal speaker comes to the syntactic processor wanting a sentence with a certain meaning; the processor's job is to complete his picture of me sentence by supplying appropriate words and phrases. The ideal hearer has a complete description of me words in me sentence but needs descriptions of the phrases and me meaning of the whole to complete the picture. A more realistic hearer starts wim a picture including imperfectly heard words and some notions about what is bei...

National statistical agencies lack statistical methodology to express uncertainty in their released estimated overall rankings. For example, the US Census Bureau produced an 'explicit' ranking of the states based on observed sample estimates during 2011 of mean travel time to work. Current literature provides measures of uncertainty in estimated individual ranks, but not a direct measure of uncertainty for the estimated overall ranking. We construct and visualize a joint confidence region for the true unknown overall ranking that provides a measure of uncertainty in the estimated overall ranking.

No abstract

Minimal residual disease (MRD) as measured by multiparameter flow cytometry (FCM) is an independent and strong prognostic factor in B-cell acute lymphoblastic leukemia (B-ALL). However, reliable flow cytometric detection of MRD strongly depends on operator skills and expert knowledge. Hence, an objective, automated tool for reliable FCM-MRD quantification, able to overcome the technical diversity and analytical subjectivity, would be most helpful. We developed a supervised machine learning approach using a combination of multiple Gaussian Mixture Models (GMM) as a parametric density model. The approach was used for finding the weights of a linear combination of multiple GMMs to represent new, "unseen" samples by an interpolation of stored samples. The experimental data set contained FCM-MRD data of 337 bone marrow samples collected at day 15 of induction therapy in three different laboratories from pediatric patients with B-ALL for which accurate, expert-set gates existed. We compared MRD quantification by our proposed GMM approach to operator assessments, its performance on data from different laboratories, as well as to other state-of-the-art automated readout methods. Our proposed GMM-combination approach proved superior over support vector machines, deep neural networks, and a single GMM approach in terms of precision and average F 1 -scores. A high correlation of expert operator-based and automated MRD assessment was achieved with reliable automated MRD quantification (F 1 -scores >0.5 in more than 95% of samples) in the clinically relevant range. Although best performance was found, if test and training samples were from the same system (i.e., flow cytometer and staining panel; lowest median F 1 -score 0.92), cross-system performance remained high with a median F 1 -score above 0.85 in all settings. In conclusion, our proposed automated approach could potentially be used to assess FCM-MRD in B-ALL in an objective and standardized manner across different laboratories.

No abstract

No abstract

No abstract

No abstract