We introduce a new compression scheme for high-dimensional vectors that approximates the vectors using sums of M codewords coming from M different codebooks. We show that the proposed scheme permits efficient distance and scalar product computations between compressed and uncompressed vectors. We further suggest vector encoding and codebook learning algorithms that can minimize the coding error within the proposed scheme. In the experiments, we demonstrate that the proposed compression can be used instead of or together with product quantization. Compared to product quantization and its optimized versions, the proposed compression approach leads to lower coding approximation errors, higher accuracy of approximate nearest neighbor search in the datasets of visual descriptors, and lower image classification error, whenever the classifiers are learned on or applied to compressed vectors.
The article describes the features of an enterprise’s business process management that concerns ad-hoc processes. The analysis of the possible implementation problems in ECM system is shown and ways of overcoming.
The article describes the features of business process management that concerns ad-hoc processes in enterprises as expert communities. The analysis of the possible implementation in corresponding Enterprise Content Management (ECM) system is shown. These results were obtained in the fourth stage of the complex project, which is carried in the frame of Government Grant with participation of NRU HSE and “IT” Corporation (Russia).
This chapter addresses the interaction between the authorities and non-state actors in HIV prevention among drug users. Based on case studies in Samara and St. Petersburg, it looks into the mix of vertical structures of governance (dominated by health authorities, including the regional AIDS centers, and drug control authorities) and aspects of network governance where both state and non-state actors collaborate on policy-making and implementation. The chapter concludes that the issue is largely left in a void outside the direct responsibility and attention of state agencies and governance networks. It is approached mainly by actors within the non-governmental sector which, for their part, do not have the resources or necessary authority to make the required policy impact.
In this article I studied the two predicative possessive constructions in Gban, which belong to the “locational” and “adnominal” formal types (lit. “A house is on John” and “John’s house exists”). They were found to be almost complementarily distributed according to which semantic types of possession they can express. The second, “adnominal”-type predicative possessive construction underwent a more detailed study. Although it has been doubted by some linguists whether there are in fact any true “adnominal” predicative possessive constructions in the languages of the world (i.e. those that are indeed based on a genuine single NP such as [John’s car]), Gban does seem to provide such a case. The “Possessor + Possessee” complex in the second predicative possessive construction here shows many properties of a single constituent and shows no differences in the syntactic behaviour from unambiguous possessive NPs in examples like ‘I saw John’s car’. At the same time, both the “Possessor + Possessee” complex and unambiguous possessive NPs in Gban only partially correspond to the theoretical expectations for an “ideal” possessive NP. They answer to these expectations in the more syntax-oriented properties, but deviate from them in the more pragmatic~semantic properties. One of these latter properties which seem unusual for NPs is the ability of both the “Possessor + Possessee” complex in the second construction and unambiguous possessive NPs to be sharply pragmatically “split”. While one part (Possessor) is fully active and topical, the other part (Possessee) can be unidentifiable and part of the focus. Cf. the possibility of questions such as, literally, “Your WHAT is there?” (‘What do you have’) or “Their WHAT did you see?” (‘What did you see of theirs?’). In the second predicative possessive construction there is also a semantic “split” — what looks and behaves syntactically as a single possessive NP here expresses, in fact, two participants of the (semantic) predicate at the same time. After that, I briefly discussed another phenomenon that demonstrates similar behaviour. There exists in Gban an alternative way of coding recipient and recipient-benefactive participants: by an adnominal NP inside the direct object NP (lit. “I bought [his clothes]” for ‘I bought clothes for him’). And in these contexts we again observe the same pragmatic “split”, and also a semantic “split”, with a single NP expressing both the undergoer and the recipient(-benefactive). To sum up, possessive noun phrases in Gban seem to have an unusual degree of pragmatic and semantic flexibility. This flexibility can be best seen in the “adnominal”-type predicative possessive construction and in the adnominal coding of recipients/recipient-benefactives. And we can also make a conjecture that probably it is this flexibility in the first place that allows such peculiar NP-based constructions to arise in a language.
This article deals with the conception of an Imperial Authority described in the “Siete Partidas” of Castilian king Alphonse the WIse (1252 – 1284) and its interpretation by a court lawyer of the Emperor of Spain Carl I (Carl V) called Gr. Lopez. The special attention is payed to the question of sovereignty, legal status of the emperor and of citizen’s right of insurrection.
We describe new features in FCART software system, an integrated environment for knowledge and data engineers with a set of research tools based on Formal Concept Analysis. The system is intended for knowledge discovery from various data sources, including structured quantitative data and text collections. Final version of data transformation from external data source into concept lattice is considered. We introduce new version of local data storage, query language for conceptual scaling of data snapshots as multi-valued contexts, and new tools for working with formal concepts.
Adyghe, a polysynthetic language of the West Caucasian family, shows the typological characteristics of ergativity, left-branching word order, and the flexibility of the lexical categories. Its word has a high degree of morphological complexity and consists of five ordered morphological zones, within which the order of affixes can vary, and recursion is possible. The information encoded in the predicate includes the argument structure, causation, and various aspectual and modal characteristics. Many meanings can be expressed, either with a combination of morphemes, or a combination of words, or with both simultaneously. There are structural asymmetries at the clause level and the principle C violations in cross-clausal syntax—the phenomenon that has been recorded also in many polysynthetic languages of America.
The chapter provides the general description of word formation in Adyghe (Northwest Caucasian). Adyghe is a highly polysynthetic language with a very weak distinction between nouns and verbs. Compounding and affixation (including both suffixation and prefixation) are widespread. Morphological means often allow recursion and the order of morphemes depends on the semantics to a large extent. Inflection and derivation are not distinguished clearly. While deverbal nominal derivation is highly developed, most “verbal” formation actually applies to all kinds of bases. Minor parts-of-speech like adjectives and adverbs show dedicated markers. Conversion proper is occasional.
This paper presents tools aimed at automating the data integration process between different SaaS-applications. Data transformations creation and customization represents a major task in this area. We propose a solution based on type theory and describe an approach to automated transformation construction in form of typed functions.
This chapter examines the personal and ideological contacts between members of the Russian émigré Eurasianist movement and representatives of the so-called “Conservative Revolution” in late Weimar Germany. Throughout the 1920s both movements professed similarly strong anti-Western and anti-democratic ideas. Yet, so far it has been little known that these groups and their members also had actual organizational contacts and direct personal interactions. Introducing new archival evidence from the Eurasianists’ personal papers, this chapter reveals that in the early 1930 Eurasianists indeed strove to forge a strategic alliance with several German rightist movements, such as “Gegner” (led by Harro Schulze-Boysen), “Die Tat” (led by Hans Zehrer), “Schwarze Front” (led by Otto Strasser) and “Widerstandsbewegung” (led by Ernst Niekisch). The Eurasianist A.P. Antipov met representatives of these groups in early February 1932, when he officially represented the Eurasianist movement at the “European Youth Congress” in Frankfurt organized by the French non-conformist Alexandre Marc and his group “Plans.” Following this event, some German “conservative revolutionaries,” in particular Schulze-Boysen, intensified their contacts and exchanged letters and programmatic statements with individual Eurasianists. By then, the Eurasianists had become part of an international, pan-European network of non-conformist groups in search of a “Third Way” between capitalism and socialism, between “left” and “right.”
In this paper, the E-pulse (extinction pulse) method known as aspect-independent ultra-wideband radar target discrimination technique is discussed. An alternative synthesis algorithm for the subsection polynomial E-pulse is introduced. The algorithm consists in building a skeleton E-pulse, its further extending and series of integration which all could be performed over the coefficients of basic functions. Not only the proposed algorithm performs up to a thousand times faster than direct matrix solution but it obtains the polynomial coefficients of the E-pulse sections avoiding the solution of a linear problem associated with ill-conditioned sparse matrix. It is proven that E-pulse signals synthesized by means of the fast algorithm and the direct one are exactly the same. To exposure the features of the E-pulse technique, two targets discriminating scheme has been simulated.
The problem of automatic detection of the moving forklift truck in video data is explored. This task is formulated in terms of computer vision approach as a moving object detection in noisy environment. It is shown that the state-of-the-art local descriptors (SURF, SIFT, FAST, ORB) are not characterized with satisfactory detection quality if the camera resolution is low, the lighting is changed dramatically and shadows are observed. In this paper we propose to use a simple mathematical morphological algorithm to detect the presence of a cargo on the forklift truck. Its first step is the estimation of the movement direction and the front part of the truck by using the updating motion history image. The second step is the application of Canny contour detection and binary morphological operations in front of the moving object to estimate simple geometric features of empty forklift. The algorithm is implemented with the OpenCV library. Our experimental study shows that the best results are achieved if the difference of the width of bounding rectangles is used as a feature. Namely, the detection accuracy is 78.7% (compare with 40% achieved by the best local descriptor), while the average frame processing time is only 5 ms (compare with 35 ms for the fastest descriptor).
An important aspect of the effectiveness of the strategy commemoration embodied in rituals, archives, virtual practices and other matters, is proposed term resonance - in response to the coherence the content and form of commemorative practices. An interesting question is how the new media, especially the Internet, resonates with the content posted on the collective memory, the last revaluation offline. So whether held online rediscovery Afghan phenomenon? In our view, there was a rather capsulisation of this phenomenon, as an escape from the resonance like a virtualization of military memory of the Afghan war.
In this article, we analyze the peculiarities in commemorative traditions of the “Afghaners,” who find it difficult to express a coherent narrative regarding their war experience. We also look at public memory about them as part of the discourse on Russian war obituaries, which contrast with the discursive customs seen in NATO obituaries for British veterans. This contrast allows an evaluation of differences in these societies’ cultural productions of public memory. The essay concludes with a reflection on the Internet’s influence on public memory regarding the Afghan war; how it gives the war a new lease on life in the digital world, yet also brings a risk of re-evaluating the war and the its’ participants actions.