Formalized reproduction of an expert‐based phytosociological classification: A case study of subalpine tall‐forb vegetation

Abstract. Delimitation of vegetation units in phytosociology is traditionally based on expert knowledge. Applications of expert‐based classifications are often inconsistent because criteria for assigning relevés to vegetation units are seldom given explicitly. Still, there is, e.g. in nature conservation, an increasing need for a consistent application of vegetation classification using computer expert systems for unit identification. We propose a procedure for formalized reproduction of an expert‐based vegetation classification, which is applicable to large phytosociological data sets. This procedure combines Bruelheide's Cocktail method with a similarity‐based assignment of relevés to constancy columns of a vegetation table. As a test of this method we attempt to reproduce the expert‐based phytosociological classification of subalpine tall‐forb vegetation of the Czech Republic which has been made by combination of expert judgement and stepwise numerical classification of 718 relevés by TWINSPAN. Applying the Cocktail method to a geographically stratified data set of 21794 relevés of all Czech vegetation types, we defined groups of species with the statistical tendency of joint occurrences in vegetation. Combinations of 12 of these species groups by logical operators AND, OR and AND NOT yielded formal definitions of 14 of 16 associations which had been accepted in the expert‐based classification. Application of these formal definitions to the original data set of 718 relevés resulted in an assignment of 376 relevés to the associations. This assignment agreed well with the original expert‐based classification. Relevés that remained un‐assigned because they had not met the requirements of any of the formal definitions, were subsequently assigned to the associations by calculating similarity to relevé groups that had already been assigned to the associations. A new index, based on frequency and fidelity, was proposed for calculating similarity. The agreement with the expert‐based classification achieved by the formal definitions was still improved after applying the similarity‐based assignment. Results indicate that the expert‐based classification can be successfully formalized and converted into a computer expert system.     

Expert system for aiding diagnosis in hearing tests]

For expert systems intended to aid diagnosis, a structure with five levels is proposed. These levels are the original area, the parameter and a reduced parameter layer, the classification and the final-decision layer. On the basis of this structures, an expert system was developed specifically for neonatal hearing screening with transitory evoked otoacoustic emissions (TEOAE). In a second step, this system was investigated for its suitability to classify emissions, regardless of patient age. For the comparison measurements in 252 mainly adult patients, some with an acquired hearing impairment, were used. To adapt the pass/fail decision to the extended evaluation criteria, the false classifications from a first run with the new data were used for training. Thereafter, the expert system, working with a wider data basis, classified the new data with a sensitivity that was increased by 4.8% to 97.2%, and a 2.0% improvement in specificity to 95.5% when classifying new data, These results, together with those of 97.3% and 94.3% achieved with exclusively neonatal TEOAE classification, clearly show the advantage of the expert system structures chosen, and document evidence of the practical applicability of the method.     

利用GIS对吉林针阔混交林TM遥感图像分类方法的初探

为提高林区TM遥感图像自动分类识别精度,在GIS技术辅助下,以吉林省汪清林业局针阔混交林TM遥感图像为例,对研究区DEM、坡向等地理因子和土壤类型等环境因子与森林植被分布之间的内在规律进行了定量分析,并结合对遥感图像预分类的定性分析,形成分类知识库,建立了适用于针阔混交林的自动分类识别专家系统.分类试验证明,该系统能比较明显地削弱混合像元和地形阴影的影响,分类精度较无监督分类法提高了14.22%,Kappa指数为0.7556,达到区别森林类型的分类目的.将GIS数据引入专家系统,应用先验知识建立推理机制,可以解决遥感图像中云区和云阴影区由于不能接收到正确的光谱值而无法进行分类的问题.     

Digital interpretation and management of land cover - a case study of Cyprus

Knowledge based methods using Geographic Information System and Remote Sensing for land cover classification and future land allocation in Cyprus are presented. Land cover was categorized combining statistical classification and expert rules, using multi-temporal Landsat TM and MSS data, and digitized maps of elevation and geology. Land needed for future infrastructure was allocated using multi-criteria evaluation. Reforestation area were identified for Pinus brutia and P. nigra comparing expert rules and multi-criteria evaluation.     

Assignment of relevés to pre‐defined classes by supervised clustering of plant communities using a new composite index

Question: How does a newly designed method of supervised clustering perform in the assignment of relevé (species composition) data to a previously established classification. How do the results compare to the assignment by experts and to the assignment using a completely different numerical method? Material: Relevés analysed represent 4186 Czech grassland plots and 4990 plots from a wide variety of vegetation types (359 different associations or basal communities) in The Netherlands. For both data sets we had at our disposal an expert classification, and for the Czech data we also had available a numerical classification as well as a classification based on a neural network method (multi‐layer perceptron). Methods: Two distance indices, one qualitative and one quantitative, are combined into a single index by weighted multiplication. The composite index is a distance index for the dissimilarity between relevés and vegetation types. For both data sets the classifications by the new method were compared with the existing classifications. Results: For the Czech grasslands we correctly classified 81% of the plots to the classes of an expert classification at the alliance level and 71% to the classes of the numerical classification. Correct classification rates for the Dutch relevés were 64, 78 and 83 % for the lowest (subassociation or association), association, and alliance level, respectively. Conclusion: Our method performs well in assigning community composition records to previously established classes. Its performance is comparable to the performance of other methods of supervised clustering. Compared with a multi‐layer perceptron (a type of artificial neural network), fewer parameters have to be estimated. Our method does not need the original relevé data for the types, but uses synoptic tables. Another practical advantage is the provision of directly interpretable information on the contributions of separate species to the result.     

Sleep classification in infants based on artificial neural networks.

The study reports on the possibility of classifying sleep stages in infants using an artificial neural network. The polygraphic data from 4 babies aged 6 weeks, 6 months and 1 year recorded over 8 hours were available for classification. From each baby 22 signals were recorded, digitized and stored on an optical disc. Subsets of these signals and additional calculated parameters were used to obtain data vectors, each of which represents an interval of 30 sec. For classification, two types of neural networks were used, a Multilayer Perceptron and a Learning Vector Quantizer. The teaching input for both networks was provided by a human expert. For the 6 sleep classes in babies aged 6 months, a 65% to 80% rate of correct classification (4 babies) was obtained for the testing data not previously seen.     

An assessment of the genetic toxicity of atrazine: Relevance to human health and environmental effects

The genetic toxicity of atrazine, a member of the s-triazine herbicides, was reviewed with the objective of classifying the chemical. Atrazine has been subjected to a broad range of genetic tests with predominantly negative results. Some publications, specifically those measuring dominant lethality in mice and bone marrow clastogenicity in rodents, reported conflicting results across two or more independent tests. Two approaches were employed to evaluate and interpret the results. The first approach attempts to classify each type of genetic endpoint as positive or negative and resolve test conflicts by critical assessment of the study and detailed data. This is the more traditional “expert judgment” approach to hazard assessment. The second approach employs a computer-assisted weight-of-evidence method of data analysis. This approach does not require resolution of conflicts but uses all data sets to arrive at a classification of hazard. The first approach was able to resolve some conflicts but not all. Use of the “expert judgement” results in an equivocal conclusion and classification. Use of the weight-of-evidence method resulted in a conclusion that atrazine does not pose a mutagenic hazard. The weight-of-evidence scheme is proposed to be a more practical and relevant approach for assessing complex data sets.     

Enzyme classification by ligand binding

The problem of assigning a biochemical function to newly discovered proteins has been traditionally approached by expert enzymological analysis, sequence analysis, and structural modeling. In recent years, the appearance of databases containing protein-ligand interaction data for large numbers of protein classes and chemical compounds have provided new ways of investigating proteins for which the biochemical function is not completely understood. In this work, we introduce a method that utilizes ligand-binding data for functional classification of enzymes. The method makes use of the existing Enzyme Commission (EC) classification scheme and the data on interactions of small molecules with enzymes from the BRENDA database. A set of ligands that binds to an enzyme with unknown biochemical function serves as a query to search a protein-ligand interaction database for enzyme classes that are known to interact with a similar set of ligands. These classes provide hypotheses of the query enzyme's function and complement other computational annotations that take advantage of sequence and structural information. Similarity between sets of ligands is computed using point set similarity measures based upon similarity between individual compounds. We present the statistics of classification of the enzymes in the database by a cross-validation procedure and illustrate the application of the method on several examples.     

Cue Recognition and Integration – Eye Tracking Evidence of Processing Differences in Sentence Comprehension in Aphasia

Purpose

We aimed at further elucidating whether aphasic patients’ difficulties in understanding non-canonical sentence structures, such as Passive or Object-Verb-Subject sentences, can be attributed to impaired morphosyntactic cue recognition, and to problems in integrating competing interpretations.

Methods

A sentence-picture matching task with canonical and non-canonical spoken sentences was performed using concurrent eye tracking. Accuracy, reaction time, and eye tracking data (fixations) of 50 healthy subjects and 12 aphasic patients were analysed.

Results

Patients showed increased error rates and reaction times, as well as delayed fixation preferences for target pictures in non-canonical sentences. Patients’ fixation patterns differed from healthy controls and revealed deficits in recognizing and immediately integrating morphosyntactic cues.

Conclusion

Our study corroborates the notion that difficulties in understanding syntactically complex sentences are attributable to a processing deficit encompassing delayed and therefore impaired recognition and integration of cues, as well as increased competition between interpretations.     

Numerical reproduction of traditional classifications and automatic vegetation identification

Questions: Is it possible to develop an expert system to provide reliable automatic identifications of plant communities at the precision level of phytosociological associations? How can unreliable expert‐based knowledge be discarded before applying supervised classification methods? Material: We used 3677 relevés from Catalonia (Spain), belonging to eight orders of terrestrial vegetation. These relevés were classified by experts into 222 low‐level units (associations or sub‐associations). Methods: We reproduced low‐level, expert‐defined vegetation units as independent fuzzy clusters using the Possibilistic C‐means algorithm. Those relevés detected as transitional between vegetation types were excluded in order to maximize the number of units numerically reproduced. Cluster centroids were then considered static and used to perform supervised classifications of vegetation data. Finally, we evaluated the classifier's ability to correctly identify the unit of both typical (i.e. training) and transitional relevés. Results: Only 166 out of 222 (75%) of the original units could be numerically reproduced. Almost all the unrecognized units were sub‐associations. Among the original relevés, 61% were deemed transitional or untypical. Typical relevés were correctly identified 95% of the time, while the efficiency of the classifier for transitional data was only 64%. However, if the second classifier's choice was also considered, the rate of correct classification for transitional relevés was 80%. Conclusions: Our approach stresses the transitional nature of relevé data obtained from vegetation databases. Relevé selection is justified in order to adequately represent the vegetation concepts associated with expert‐defined units.     

Phenotype Classification of Zebrafish Embryos by Supervised Learning

Zebrafish is increasingly used to assess biological properties of chemical substances and thus is becoming a specific tool for toxicological and pharmacological studies. The effects of chemical substances on embryo survival and development are generally evaluated manually through microscopic observation by an expert and documented by several typical photographs. Here, we present a methodology to automatically classify brightfield images of wildtype zebrafish embryos according to their defects by using an image analysis approach based on supervised machine learning. We show that, compared to manual classification, automatic classification results in 90 to 100% agreement with consensus voting of biological experts in nine out of eleven considered defects in 3 days old zebrafish larvae. Automation of the analysis and classification of zebrafish embryo pictures reduces the workload and time required for the biological expert and increases the reproducibility and objectivity of this classification.     

The Willingness of Children to Lie and the Assessment of Credibility in an Ecologically Relevant Laboratory Setting

A series of 4 experiments were conducted to (a) determine the willingness of children to lie in a realistic setting, (b) compare judgments of credibility from both lay evaluations and Criterion-Based Content Analysis (CBCA), and (c) examine the effects of expert testimony regarding Statement Validity Assessment on mock jurors who were asked to make evaluations of the children's statements. In Experiment 1, 81% of children who witnessed a research assistant steal a textbook made accusations against the thief (truthful), 69% of children who did not witness the theft accused the research assistant of the theft following prompting by significant others, and 56% of the children who witnessed a significant other steal the textbook incorrectly accused the research assistant following a request from their significant other. Using the statements obtained from the children in Experiment 1, Experiments 2 and 3 found that classification accuracy of lay evaluators was significantly poorer than expert application of CBCA, which resulted in 89% classification accuracy. Finally, brief exposure to CBCA expert testimony appeared to have no benefits on mock jurors' assessments of credibility of the children's statements in Experiment 4. Implications for North American legal systems are discussed.     

Constructing a fuzzy rule-based system using the ILFN network and Genetic Algorithm.

In this paper, a method for automatic construction of a fuzzy rule-based system from numerical data using the Incremental Learning Fuzzy Neural (ILFN) network and the Genetic Algorithm is presented. The ILFN network was developed for pattern classification applications. The ILFN network, which employed fuzzy sets and neural network theory, equips with a fast, one-pass, on-line, and incremental learning algorithm. After trained, the ILFN network stored numerical knowledge in hidden units, which can then be directly interpreted into if then rule bases. However, the rules extracted from the ILFN network are not in an optimized fuzzy linguistic form. In this paper, a knowledge base for fuzzy expert system is extracted from the hidden units of the ILFN classifier. A genetic algorithm is then invoked, in an iterative manner, to reduce number of rules and select only discriminate features from input patterns needed to provide a fuzzy rule-based system. Three computer simulations using a simulated 2-D 3-class data, the well-known Fisher's Iris data set, and the Wisconsin breast cancer data set were performed. The fuzzy rule-based system derived from the proposed method achieved 100% and 97.33% correct classification on the 75 patterns for training set and 75 patterns for test set, respectively. For the Wisconsin breast cancer data set, using 400 patterns for training and 299 patterns for testing, the derived fuzzy rule-based system achieved 99.5% and 98.33% correct classification on the training set and the test set, respectively.     

Animal Sound Identifier (ASI): software for automated identification of vocal animals

Automated audio recording offers a powerful tool for acoustic monitoring schemes of bird, bat, frog and other vocal organisms, but the lack of automated species identification methods has made it difficult to fully utilise such data. We developed Animal Sound Identifier (ASI), a MATLAB software that performs probabilistic classification of species occurrences from field recordings. Unlike most previous approaches, ASI locates training data directly from the field recordings and thus avoids the need of pre‐defined reference libraries. We apply ASI to a case study on Amazonian birds, in which we classify the vocalisations of 14 species in 194 504 one‐minute audio segments using in total two weeks of expert time to construct, parameterise, and validate the classification models. We compare the classification performance of ASI (with training templates extracted automatically from field data) to that of monitoR (with training templates extracted manually from the Xeno‐Canto database), the results showing ASI to have substantially higher recall and precision rates.     

Dimensionality reduction for visualization of normal and pathological speech data

For an adequate analysis of pathological speech signals, a sizeable number of parameters is required, such as those related to jitter, shimmer and noise content. Often this kind of high-dimensional signal representation is difficult to understand, even for expert voice therapists and physicians. Data visualization of a high-dimensional dataset can provide a useful first step in its exploratory data analysis, facilitating an understanding about its underlying structure. In the present paper, eight dimensionality reduction techniques, both classical and recent, are compared on speech data containing normal and pathological speech. A qualitative analysis of their dimensionality reduction capabilities is presented. The transformed data are also quantitatively evaluated, using classifiers, and it is found that it may be advantageous to perform the classification process on the transformed data, rather than on the original. These qualitative and quantitative analyses allow us to conclude that a nonlinear, supervised method, called kernel local Fisher discriminant analysis is superior for dimensionality reduction in the actual context.     

A Realization Optimality-Theoretic approach to affix order

The interplay of the main factors affecting affix order in inflection (semantic scope, phonology, and morphological templates) can be accounted for in an inferential-realizational Optimality-Theoretic model of morphology, which we present here. Within this model, phonological form is spelled out by means of individual-language-particular realization constraints that associate abstract morphosyntactic feature values with phonological forms and that are ordered among more general constraints governing factors like scope and feature splitting. The data used to exemplify the application of our theory to affix order are drawn from Haspelmath’s (A grammar of Lezgian, Mouton de Gruyter, Berlin, 1993) grammar of Lezgian, a language of the Northeast Caucasian family spoken largely in Dagestan (Russia) and Azerbaijan.     

Effects of Grammaticality and Morphological Complexity on the P600 Event-Related Potential Component

We investigated interactions between morphological complexity and grammaticality on electrophysiological markers of grammatical processing during reading. Our goal was to determine whether morphological complexity and stimulus grammaticality have independent or additive effects on the P600 event-related potential component. Participants read sentences that were either well-formed or grammatically ill-formed, in which the critical word was either morphologically simple or complex. Results revealed no effects of complexity for well-formed stimuli, but the P600 amplitude was significantly larger for morphologically complex ungrammatical stimuli than for morphologically simple ungrammatical stimuli. These findings suggest that some previous work may have inadequately characterized factors related to reanalysis during morphosyntactic processing. Our results show that morphological complexity by itself does not elicit P600 effects. However, in ungrammatical circumstances, overt morphology provides a more robust and reliable cue to morphosyntactic relationships than null affixation.     

A bioturbation classification of European marine infaunal invertebrates

Bioturbation, the biogenic modification of sediments through particle reworking and burrow ventilation, is a key mediator of many important geochemical processes in marine systems. In situ quantification of bioturbation can be achieved in a myriad of ways, requiring expert knowledge, technology, and resources not always available, and not feasible in some settings. Where dedicated research programmes do not exist, a practical alternative is the adoption of a trait‐based approach to estimate community bioturbation potential (BP_c). This index can be calculated from inventories of species, abundance and biomass data (routinely available for many systems), and a functional classification of organism traits associated with sediment mixing (less available). Presently, however, there is no agreed standard categorization for the reworking mode and mobility of benthic species. Based on information from the literature and expert opinion, we provide a functional classification for 1033 benthic invertebrate species from the northwest European continental shelf, as a tool to enable the standardized calculation of BP_c in the region. Future uses of this classification table will increase the comparability and utility of large‐scale assessments of ecosystem processes and functioning influenced by bioturbation (e.g., to support legislation). The key strengths, assumptions, and limitations of BP_c as a metric are critically reviewed, offering guidelines for its calculation and application.     

The ability of video image analysis to predict lean meat yield and EUROP score of lamb carcasses

A total of 862 lamb carcasses that were evaluated by both the VIAscan® and the current EUROP classification system were deboned and the actual yield was measured. Models were derived for predicting lean meat yield of the legs (Leg%), loin (Loin%) and shoulder (Shldr%) using the best VIAscan® variables selected by stepwise regression analysis of a calibration data set (n=603). The equations were tested on validation data set (n=259). The results showed that the VIAscan® predicted lean meat yield in the leg, loin and shoulder with an R² of 0.60, 0.31 and 0.47, respectively, whereas the current EUROP system predicted lean yield with an R² of 0.57, 0.32 and 0.37, respectively, for the three carcass parts. The VIAscan® also predicted the EUROP score of the trial carcasses, using a model derived from an earlier trial. The EUROP classification from VIAscan® and the current system were compared for their ability to explain the variation in lean yield of the whole carcass (LMY%) and trimmed fat (FAT%). The predicted EUROP scores from the VIAscan® explained 36% of the variation in LMY% and 60% of the variation in FAT%, compared with the current EUROP system that explained 49% and 72%, respectively. The EUROP classification obtained by the VIAscan® was tested against a panel of three expert classifiers (n=696). The VIAscan® classification agreed with 82% of conformation and 73% of the fat classes assigned by a panel of expert classifiers. It was concluded that VIAscan® provides a technology that can directly predict LMY% of lamb carcasses with more accuracy than the current EUROP classification system. The VIAscan® is also capable of classifying lamb carcasses into EUROP classes with an accuracy that fulfils minimum demands for the Icelandic sheep industry. Although the VIAscan® prediction of the Loin% is low, it is comparable to the current EUROP system, and should not hinder the adoption of the technology to estimate the yield of Icelandic lambs as it delivered a more accurate prediction for the Leg%, Shldr% and overall LMY% with negligible prediction bias.