Place categorization and semantic mapping on a mobile robot

In this paper we focus on the challenging problem of place categorization and semantic mapping on a robot with-out environment-specific training. Motivated by their ongoing success in various visual recognition tasks, we build our system upon a state-of-the-art convolutional network. We overcome its closed-set limitations by complementing the network with a series of one-vs-all classifiers that can learn to recognize new semantic classes online. Prior domain knowledge is incorporated by embedding the classification system into a Bayesian filter framework that also ensures temporal coherence. We evaluate the classification accuracy of the system on a robot that maps a variety of places on our campus in real-time. We show how semantic information can boost robotic object detection performance and how the semantic map can be used to modulate the robot’s behaviour during navigation tasks. The system is made available to the community as a ROS module.

Long-term effects of physical activity level on changes in healthy body mass index over 12 years in young adult women

Objectives To examine the effects of overall level and timing of physical activity (PA) on changes from a healthy body mass index (BMI) category over 12 years in young adult women. Patients and Methods Participants in the Australian Longitudinal Study on Women's Health (younger cohort, born 1973-1978) completed surveys between 2000 (age 22-27 years) and 2012 (age 34-39 years). Physical activity was measured in 2000, 2003, 2006, and 2009 and was categorized as very low, low, active, or very active at each survey, and a cumulative PA score for this 9-year period was created. Logistic regression was used to examine relationships between PA accumulated across all surveys (cumulative PA model) and PA at each survey (critical periods PA model), with change in BMI category (from healthy to overweight or healthy to obese) from 2000 to 2012. Results In women with a healthy BMI in 2000, there were clear dose-response relationships between accumulated PA and transition to overweight (P=.03) and obesity (P<.01) between 2000 and 2012. The critical periods analysis indicated that very active levels of PA at the 2006 survey (when the women were 28-33 years old) and active or very active PA at the 2009 survey (age 31-36 years) were most protective against transitioning to overweight and obesity. Conclusion These findings confirm that maintenance of very high PA levels throughout young adulthood will significantly reduce the risk of becoming overweight or obese. There seems to be a critical period for maintaining high levels of activity at the life stage when many women face competing demands of caring for infants and young children.

Spectral unmixing with estimated adaptive endmember index using Extended Support Vector Machine

The most difficult operation in the flood inundation mapping using optical flood images is to separate fully inundated areas from the ‘wet’ areas where trees and houses are partly covered by water. This can be referred as a typical problem the presence of mixed pixels in the images. A number of automatic information extraction image classification algorithms have been developed over the years for flood mapping using optical remote sensing images. Most classification algorithms generally, help in selecting a pixel in a particular class label with the greatest likelihood. However, these hard classification methods often fail to generate a reliable flood inundation mapping because the presence of mixed pixels in the images. To solve the mixed pixel problem advanced image processing techniques are adopted and Linear Spectral unmixing method is one of the most popular soft classification technique used for mixed pixel analysis. The good performance of linear spectral unmixing depends on two important issues, those are, the method of selecting endmembers and the method to model the endmembers for unmixing. This paper presents an improvement in the adaptive selection of endmember subset for each pixel in spectral unmixing method for reliable flood mapping. Using a fixed set of endmembers for spectral unmixing all pixels in an entire image might cause over estimation of the endmember spectra residing in a mixed pixel and hence cause reducing the performance level of spectral unmixing. Compared to this, application of estimated adaptive subset of endmembers for each pixel can decrease the residual error in unmixing results and provide a reliable output. In this current paper, it has also been proved that this proposed method can improve the accuracy of conventional linear unmixing methods and also easy to apply. Three different linear spectral unmixing methods were applied to test the improvement in unmixing results. Experiments were conducted in three different sets of Landsat-5 TM images of three different flood events in Australia to examine the method on different flooding conditions and achieved satisfactory outcomes in flood mapping.

Explorations in the distributional and semantic similarity of words

A straightforward computation of the list of the words (the `tail words' of the list) that are distributionally most similar to a given word (the `head word' of the list) leads to the question: How semantically similar to the head word are the tail words; that is: how similar are their meanings to its meaning? And can we do better? The experiment was done on nearly 18,000 most frequent nouns in a Finnish newsgroup corpus. These nouns are considered to be distributionally similar to the extent that they occur in the same direct dependency relations with the same nouns, adjectives and verbs. The extent of the similarity of their computational representations is quantified with the information radius. The semantic classification of head-tail pairs is intuitive; some tail words seem to be semantically similar to the head word, some do not. Each such pair is also associated with a number of further distributional variables. Individually, their overlap for the semantic classes is large, but the trained classification-tree models have some success in using combinations to predict the semantic class. The training data consists of a random sample of 400 head-tail pairs with the tail word ranked among the 20 distributionally most similar to the head word, excluding names. The models are then tested on a random sample of another 100 such pairs. The best success rates range from 70% to 92% of the test pairs, where a success means that the model predicted my intuitive semantic class of the pair. This seems somewhat promising when distributional similarity is used to capture semantically similar words. This analysis also includes a general discussion of several different similarity formulas, arranged in three groups: those that apply to sets with graded membership, those that apply to the members of a vector space, and those that apply to probability mass functions.

Three-year-olds are sensitive to semantic prominence during online language comprehension

Recent evidence from adult pronoun comprehension suggests that semantic factors such as verb transitivity affect referent salience and thereby anap- hora resolution. We tested whether the same semantic factors influence pronoun comprehension in young children. In a visual world study, 3-year- olds heard stories that began with a sentence containing either a high or a low transitivity verb. Looking behaviour to pictures depicting the subject and object of this sentence was recorded as children listened to a subsequent sentence containing a pronoun. Children showed a stronger preference to look to the subject as opposed to the object antecedent in the low transitivity condition. In addition there were general preferences (1) to look to the subject in both conditions and (2) to look more at both potential antecedents in the high transitivity condition. This suggests that children, like adults, are affected by semantic factors, specifically semantic prominence, when interpreting anaphoric pronouns.

The connection between on-farm assessed welfare (A-index) and production parameters of the sows

Tämän tutkimuksen tavoitteena oli selvittää tilalla määritetyn hyvinvoinnin yhteyttä emakoiden tuotantotuloksiin. Hyvinvointia arvioitiin suomalaisen hyvinvointi-indeksin, A-indeksi, avulla. Tuotantotuloksina käytettiin kahta erilaista tuotosaineistoa, jotka molemmat pohjautuivat kansalliseen tuotosseuranta aineistoon. Hyvinvointimääritykset tehtiin 30 porsastuotantosikalassa maaliskuun 2007 aikana. A-indeksi koostuu kuudesta kategoriasta ’liikkumismahdollisuudet’, ’alustan ominaisuudet’, ’sosiaaliset kontaktit’, ’valo, ilma ja melu’, ’ruokinta ja veden saanti’ sekä ’eläinten terveys ja hoidon taso’. Jokaisessa kategoriassa on 3-10 pääosin ympäristöperäistä muuttujaa, jotka vaihtelevat osastoittain. Maksimipistemäärä osastolle on 100. Hyvinvointimittaukset tehtiin porsitus-, tiineytys- ja joutilasosastoilla. Erillisten tiineytysosastojen pienen lukumäärän takia (n=7) tilakohtaiset tiineytys- ja joutilasosastopisteet yhdistettiin ja keskiarvoja käytettiin analyyseissä. Yhteyksiä tuotokseen tutkittiin kahden eri aineiston avulla 1) Tilaraportti aineisto (n=29) muodostuu muokkaamattomista tila- ja tuotostuloksista tilavierailua edeltävän vuoden ajalta, 2) POTSIaineisto (n=30) muodostuu POTSI-ohjelmalla (MTT) muokatusta tuotantoaineistosta, joka sisältää managementtiryhmän (tila, vuosi, vuodenaika) vaikutuksen ensikoiden ja emakoiden pahnuekohtaiseen tuotokseen. Yhteyksiä analysointiin korrelaatio- ja regressioanalyysien avulla. Vaikka osallistuminen tutkimukseen oli vapaaehtoista, molempien tuotantoaineistojen perusteella tutkimustilat edustavat keskituottoista suomalaista sikatilaa. A-indeksin kokonaispisteet vaihtelivat välillä 37,5–64,0 porsitusosastolla ja 39,5–83,5 joutilasosastolla. Tilaraporttiaineistoa käytettäessä paremmat pisteet porsitusosaston ’eläinten terveys ja hoidon taso’ -kategoriasta lyhensivät eläinten lisääntymissykliä, lisäsivät syntyvien pahnueiden ja porsaiden määrää sekä alensivat kuolleena syntyneiden lukumäärää. Regressiomallin mukaan ’eläinten terveys ja hoidon taso’ -kategoria selitti syntyvien porsaiden lukumäärän, porsimisvälin pituuden sekä keskiporsimiskerran vaihtelua. Paremmat pisteet joutilasosaston ’liikkumismahdollisuudet’ kategoriasta alensivat syntyneiden pahnueiden sekä syntyneiden että vieroitettujen porsaiden lukumäärää. Regressiomallin mukaan ensikkopahnueiden osuus ja ”liikkumismahdollisuudet” kategorian pisteet selittivät vieroitettujen porsaiden lukumäärän vaihtelua. POTSI-aineiston yhteydessä kuolleena syntyneiden porsaiden lukumäärän aleneminen oli ensikoilla yhteydessä parempiin porsitusosaston ’sosiaalisiin kontakteihin’ ja emakoilla puolestaan joutilasosaston parempiin ’eläinten terveys ja hoidon taso’ pisteisiin. Kahden eri tuotantoaineiston avulla saadut tulokset erosivat toisistaan. Seuraavissa tutkimuksissa onkin suositeltavampaa käyttää Tilaraporttiaineistoja, joissa tuotokset ilmoitetaan vuosikohtaisina. Tämän tutkimuksen perusteella hyvinvoinnilla ja tuotoksella on yhteyksiä, joilla on myös merkittävää taloudellista vaikutusta. Erityisesti hyvä eläinten hoito ja eläinten terveys lisäävät tuotettujen porsaiden määrää ja lyhentävät lisääntymiskiertoa. Erityishuomiota tulee kiinnittää vapaana olevien joutilaiden emakoiden sosiaaliseen stressiin ja rehunsaannin varmistamiseen kaikille yksilöille.

Choosing an appropriate index to construct dominance hierarchies in animal societies: a comparison of three indices

A plethora of indices have been proposed and used to construct dominance hierarchies in a variety of vertebrate and invertebrate societies, although the rationale for choosing a particular index for a particular species is seldom explained. In this study, we analysed and compared three such indices, viz Clutton-Brock et al.'s index (CBI), originally developed for red deer, Cervus elaphus, David's score (DS) originally proposed by the statistician H. A. David and the frequency-based index of dominance (FDI) developed and routinely used by our group for the primitively eusocial wasps Ropalidia marginata and Ropalidia cyathiformis. Dominance ranks attributed by all three indices were strongly and positively correlated for both natural data sets from the wasp colonies and for artificial data sets generated for the purpose. However, the indices differed in their ability to yield unique (untied) ranks in the natural data sets. This appears to be caused by the presence of noninteracting individuals and reversals in the direction of dominance in some of the pairs in the natural data sets. This was confirmed by creating additional artificial data sets with noninteracting individuals and with reversals. Based on the criterion of yielding the largest proportion of unique ranks, we found that FDI is best suited for societies such as the wasps belonging to Ropalidia, DS is best suited for societies with reversals and CBI remains a suitable index for societies such as red deer in which multiple interactions are uncommon. (C) 2009 The Association for the Study of Animal Behaviour. Published by Elsevier Ltd. All rights reserved.

Structural aspects of semantic-directed clusters

Knowledge-based clusters are studied from the structural point of view. Generalized descriptions for such clusters are stated and illustrated. Peculiarities of certain knowledge-based cluster configurations are highlighted. The adequacy of the connectives logical and (“and”) logical or (“exclusive-or”) in describing such clusters is justified. The definition of “concept” is elaborated from the clustering point of view and used to establish the equivalence between, descriptions of clusters and concepts. The order-independence of semantic-directed clustering approach is established formally based on axiomatic considerations.

Maximum compatible classes from compatibility matrices

A fast algorithm for the computation of maximum compatible classes (mcc) among the internal states of an incompletely specified sequential machine is presented in this paper. All the maximum compatible classes are determined by processing compatibility matrices of progressingly diminishing order, whose total number does not exceed (p + m), where p is the largest cardinality among these classes, and m is the number of such classes. Consequently the algorithm is specially suitable for the state minimization of very large sequential machines as encountered in vlsi circuits and systems.

The expiration day effect of indexoptions and index futures on the underlying shares

This paper studies the effect of the expiration day of index options and futures on the trading volume, variance and price of the underlying shares. The data consists of all trades for the underlying shares in the FOX-index for expiration days during the period October 1995 to the mid of yer 1999. The main results seem to support the findings of Kan 2001, i.e. no manipulation on a larger scale. However, some indication of manipulation could be found if certain characteristics are favorable. These characteristics include: a) a large quantity of outstanding futures or at/in the money options contracts, b) there exists shares with high index weight but fairly low trading volume. Lastly, there is some indication that manipulation might be more popular towards the end of the examined time period.

Pricing Index Options with Stochastic Volatility

The objective of this paper is to investigate the pricing accuracy under stochastic volatility where the volatility follows a square root process. The theoretical prices are compared with market price data (the German DAX index options market) by using two different techniques of parameter estimation, the method of moments and implicit estimation by inversion. Standard Black & Scholes pricing is used as a benchmark. The results indicate that the stochastic volatility model with parameters estimated by inversion using the available prices on the preceding day, is the most accurate pricing method of the three in this study and can be considered satisfactory. However, as the same model with parameters estimated using a rolling window (the method of moments) proved to be inferior to the benchmark, the importance of stable and correct estimation of the parameters is evident.

The Physical Strenuousness of Work is Slightly Associated with an Upward Trend in the Body Mass Index

This paper explores the relationship between the physical strenuousness of work and the body mass index in Finland, using individual microdata over the period 1972-2002. The data contain self-reported information about the physical strenuousness of a respondent’s current occupation. Our estimates show that the changes in the physical strenuousness of work can explain around 8% at most of the definite increase in BMI observed over the period. The main reason for this appears to be that the quantitative magnitude of the effect of the physical strenuousness of work on BMI is rather moderate. Hence, according to the point estimates, BMI is only around 1.5% lower when one’s current occupation is physically very demanding and involves lifting and carrying heavy objects compared with sedentary job (reference group of the estimations), other things being equal. Accordingly, the changes in eating habits and the amount of physical activity during leisure time must be the most important contributors to the upward trend in BMI in industrialised countries, but not the changes in the labour market structure.