119,060 research outputs found
Reason to Explain: Interactive Contrastive Explanations (REASONX)
Many high-performing machine learning models are not interpretable. As they are increasingly used in decision scenarios that can critically affect individuals, it is necessary to develop tools to better understand their outputs. Popular explanation methods include contrastive explanations. However, they suffer several shortcomings, among others an insufficient incorporation of background knowledge, and a lack of interactivity. While (dialogue-like) interactivity is important to better communicate an explanation, background knowledge has the potential to significantly improve their quality, e.g., by adapting the explanation to the needs of the end-user. To close this gap, we present reasonx, an explanation tool based on Constraint Logic Programming (CLP). reasonx provides interactive contrastive explanations that can be augmented by background knowledge, and allows to operate under a setting of under-specified information, leading to increased flexibility in the provided explanations. reasonx computes factual and contrastive decision rules, as well as closest contrastive examples. It provides explanations for decision trees, which can be the ML models under analysis, or global/local surrogate models of any ML model. While the core part of reasonx is built on CLP, we also provide a program layer that allows to compute the explanations via Python, making the tool accessible to a wider audience. We illustrate the capability of reasonx on a synthetic data set, and on a well-developed example in the credit domain. In both cases, we can show how reasonx can be flexibly used and tailored to the needs of the user
Declarative Reasoning on Explanations Using Constraint Logic Programming
Explaining opaque Machine Learning (ML) models is an increasingly relevant problem. Current explanation in AI (XAI) methods suffer several shortcomings, among others an insufficient incorporation of background knowledge, and a lack of abstraction and interactivity with the user. We propose reasonx, an explanation method based on Constraint Logic Programming (CLP). reasonx can provide declarative, interactive explanations for decision trees, which can be the ML models under analysis or global/local surrogate models of any black-box model. Users can express background or common sense knowledge using linear constraints and MILP optimization over features of factual and contrastive instances, and interact with the answer constraints at different levels of abstraction through constraint projection. We present here the architecture of reasonx, which consists of a Python layer, closer to the user, and a CLP layer. reasonx’s core execution engine is a Prolog meta-program with declarative semantics in terms of logic theories
Time in a Multi-Theory Logical Framework
We present a knowledge representation framework where a collection of logic programs can be combined together by means of meta-level program composition operations. Each object-level program is composed by a collection of extended clauses, equipped with a time interval representing the time period in which they hold. The interaction between program composition operations and time yields a powerful knowledge representation language in which many applications can be naturally developed. The language is given a meta-level semantics which provides also an executable specification. Two applications in the field of business regulations are show
Going Beyond Counting First Authors in Author Co-citation Analysis
The present study examines one of the fundamental aspects of author co-citation analysis (ACA) - the way co-citation
counts are defined. Co-citation counting provides the data on which all subsequent statistical analyses and mappings
are based, and we compare ACA results based on two different types of co-citation counting - the traditional type that
only counts the first one among a cited work's authors on the one hand and a non-traditional type that takes into
account the first 5 authors of a cited work on the other hand. Results indicate that the picture produced through this non-traditional author co-citation counting contains more coherent author groups and is therefore considerably clearer. However, this picture represents fewer specialties in the research field being studied than that produced through the traditional first-author co-citation counting when the same number of top-ranked authors is selected and analyzed. Reasons for these effects are discussed
Oral administration of chestnut tannins to reduce the duration of neonatal calf diarrhea
Background: Neonatal calf diarrhea is generally caused by infectious agents and is a very common disease
in bovine practice, leading to substantial economic losses. Tannins are known for their astringent and anti- inflammatory properties in the gastro-enteric tract. The aim of this study was to evaluate the effect of the oral administration of chestnut tannins (Castanea sativa Mill.) in order to reduce the duration of calf neonatal diarrhea. Twenty-four Italian Friesian calves affected by neonatal diarrhea were included. The duration of the diarrheic episode (DDE) was recorded and the animals were divided into a control group (C), which received Effydral® in
2 l of warm water, and a tannin-treated group (T), which received Effydral® in 2 l of warm water plus 10 g of extract of chestnut tannins powder. A Mann-Whitney test was performed to verify differences for the DDE values between the two groups.
Results: The DDE was significantly higher in group C than in group T (p = 0.02), resulting in 10.1 ± 3.2 and 6.6 ± 3. 8 days, respectively.
Conclusions: Phytotherapic treatments for various diseases have become more common both in human and in veterinary medicine, in order to reduce the presence of antibiotic molecules in the food chain and in the environment. Administration of tannins in calves with diarrhea seemed to shorten the DDE in T by almost 4 days compared to C, suggesting an effective astringent action of chestnut tannins in the calf, as already reported in humans. The use of chestnut tannins in calves could represent an effective, low-impact treatment for neonatal diarrhea
Linseed supplementation in the diet of fattening pigs: Effect on the fatty acid profile of different pork cuts
Linseed supplementation improves the fatty acid (FA) profile of meat, primarily by increasing α-linolenic acid (ALA). Pork is one of the most consumed meats, though the high level of saturated FA, therefore, a change of the FA profile is needed to improve the health properties. The aim of this work was to study the influence of extruded linseed supplementation on the FA profile in 5 different pork cuts, increasing their nutraceutical properties. Sixty pigs were divided into two groups, which were fed two different diets: control (C) and experimental (L, 8% supplementation with extruded flaxseed). Five cuts were sampled: backfat (Bf), bacon (B), Boston shoulder (Bs), ham lean part (Hl) and ham fatty part (Hf). The L diet reduced fat content in Hf (−6%) and B (−11%), while the other cuts did not show differences. Moreover, L group showed a higher level of n-3 PUFA (approx. 9-fold), with a significant reduction in the n-6/n-3 from 20 to 2.5. In cuts rich in fat (Bf, B and Hf) of L group, the level of n-3 PUFA was higher than the thresholds set by EU to obtain the claim “Source of omega-3 fatty acids”. In contrast, the lean cuts (Hl and Bs) did not exceed the threshold of n-3 PUFA required to obtain the claim, as consequence of the low level of fat. The results demonstrated that a diet with 8% extruded linseed enhanced the nutraceutical quality of pork meat
Square Dancing with the Stars to Enhance Dynamic Hirschman Linkages?
In this Presidential Address, the author takes the reader on a reconnaissance of his life and time as a regional scientist. He points out scenery he found scintillating along the way, hoping that some may pick up the banner and chew on a few of the ideas for a while. He suggests a revisit to Albert O. Hirschman’s notion of key sectors and more empirical analysis related to Marcus Berliant’s and Masahisa Fujita’s notion of knowledge creation and transfer.Presidential Address, San Antonio, Texas, March 29, 2014 (53rd Meetings of the Southern Regional Science Association
Appropriate Similarity Measures for Author Cocitation Analysis
We provide a number of new insights into the methodological discussion about author cocitation analysis. We first argue that the use of the Pearson correlation for measuring the similarity between authors’ cocitation profiles is not very satisfactory. We then discuss what kind of similarity measures may be used as an alternative to the Pearson correlation. We consider three similarity measures in particular. One is the well-known cosine. The other two similarity measures have not been used before in the bibliometric literature. Finally, we show by means of an example that our findings have a high practical relevance.information science;Pearson correlation;cosine;similarity measure;author cocitation analysis
Radio-guided surgery in recurrent renal hyperparathyroidism: Report of a case
BACKGROUND: It has been demonstrated that radio-guided surgery offers several advantages in treating primary hyperparathyroidism. Even if it is considered less helpful in renal hyperparathyroidism, it could be of tremendous advantage in the treatment of persistent or recurrent secondary hyperparathyroidism. METHODS: We report a case of recurrent secondary hyperparathyroidism treated by the use of radio-guided surgery. The preoperative assessment consisting of ultrasonography, magnetic resonance imaging, and 99mTc-sestamibi scintigraphy identified a parathyroid in the upper mediastinum. The patient underwent a radio-guided neck re-exploration that allowed a rapid localization and excision of the ectopic gland, which was located in the anterosuperior mediastinum, in front of the trachea, between the innominant and the left common carotid artery. RESULTS: The operative time was 45 minutes. The patient was discharged on the first postoperative day. A decrease in serum calcium and parathyroid hormone was observed subsequently. A follow-up of 6 months did not show any recurrence. CONCLUSIONS: The case reported indicates that radio-guided surgery can help surgeons detect parathyroid tissue in selected cases of renal hyperparathyroidism
- …
