Label noise is an important issue in classification, with many potential negative consequences. For example, the accuracy of predictions may decrease, whereas the complexity of inferred models and the number of necessary training samples may increase. Many works in the literature have been devoted to the study of label noise and the development of techniques to deal with label noise. However, the field lacks a comprehensive survey on the different types of label noise, their consequences and the algorithms that consider label noise. This paper proposes to fill this gap. First, the definitions and sources of label noise are considered and a taxonomy of the types of label noise is proposed. Second, the potential consequences of label noise are discussed. Third, label noise-robust, label noise cleansing, and label noise-tolerant algorithms are reviewed. For each category of approaches, a short discussion is proposed to help the practitioner to choose the most suitable technique in its own particular field of application. Eventually, the design of experiments is also discussed, what may interest the researchers who would like to test their own algorithms. In this paper, label noise consists of mislabeled instances: no additional information is assumed to be available like e.g., confidences on labels.

https://romisatriawahono.net/lecture/rm/survey/machine%20learning/Frenay%20-%20Classification%20in%20the%20Presence%20of%20Label%20Noise%20-%202014.pdf

Classification in the Presence of Label Noise: A Survey

https://6000.co.za/wp-content/uploads/2013/02/2013-Cawthorn.pdf

A high incidence of species substitution and mislabelling detected in meat products sold in South Africa

Following a request from the Food Safety Authority of Ireland, the EFSA Panel on Dietetic Products, Nutrition and Allergies (NDA Panel) was asked to deliver a scientific opinion on the evaluation of allergenic foods and food ingredients for labelling purposes In view of the request, the NDA Panel decided to update its previous opinions relative to food ingredients or substances with known allergenic potential listed in Annex IIIa of 2003/89/EC, as amended These include cereals containing gluten, milk and dairy products, eggs, nuts, peanuts, soy, fish, crustaceans, molluscs, celery, lupin, sesame, mustard and sulphites The opinion relates to immunoglobulin (Ig)E- and non-IgE-mediated food allergy, to coeliac disease and to adverse reactions to sulphites in food, and it does not address non-immune-mediated adverse reactions to food It includes information on the prevalence of food allergy in unselected populations, proteins identified as food allergens, cross-reactivities, the effects of food processing on the allergenicity of foods and ingredients, methods for the detection of allergens and allergenic foods, doses observed to trigger adverse reactions in sensitive individuals and risk assessment methodologies that have been used to derive individual and population thresholds for selected allergenic foods

Scientific Opinion on the evaluation of allergenic foods and food ingredients for labelling purposes

Authenticity of dairy products has become an urgent issue for producers, researchers, governments, consumers and so on due to the increase of falsification procedures inducing lost large of money as well as the confidence of consumers. The determination of the authenticity and the detection of adulteration of milk and dairy products have been determined by several analytical techniques (e.g., physico-chemical, sensory, chromatography, and so on). Although these methods are considered as the reference ones, they required sophisticated analytical equipment's and skilled operators; they are also time consuming and need both the purchase and disposal of chemical reagents. Therefore, there is a need to find cheap and fast methods for the determination of the authenticity and the detection of adulteration of these products. Thus, spectroscopic techniques such as fluorescence spectroscopy, near infrared (NIR), mid infrared (MIR), nuclear magnetic resonance (NMR), among others, in combination with multivariate data analysis methods could be considered helpful tools in this domain. The advantages and disadvantages of each technique will be also discussed in this review.

Analytical methods coupled with chemometric tools for determining the authenticity and detecting the adulteration of dairy products: A review

DNA-based techniques for authentication of processed food and food supplements.

DNA barcoding reveals a high incidence of fish species misrepresentation and substitution on the South African market

Evaluation of the 16S and 12S rRNA genes as universal markers for the identification of commercial fish species in South Africa.

Comparative study of different methods for the extraction of DNA from fish species commercially available in South Africa

The limitations intrinsic to morphology-based identification systems have created an urgent need for reliable genetic methods that enable the unequivocal recognition of fish species, particularly those that are prone to overexploitation and/or market substitution. The aim of this study was to develop a comprehensive reference library of DNA sequence data to allow the explicit identification of 53 commercially available fish species in South Africa, most of which were locally caught marine species. Sequences of approximately 655 base pairs were generated for all species from the cytochrome c oxidase I (COI) gene, the region widely adopted for DNA barcoding. Specimens of the genus Thunnus were examined in further detail, employing additional mitochondrial DNA control region sequencing. Cumulative analysis of the sequences from the COI region revealed mean conspecific, congeneric and confamilial Kimura 2-parameter distances of 0.10%, 4.58% and 15.43%, respectively. The results showed that the vast majority (98%) of fish species examined could be readily differentiated by their COI barcodes, but that supplementary control region sequencing was more useful for the discrimination of three Thunnus species. Additionally, the analysis of COI data raised the prospect that Thyrsites atun (snoek) could constitute a species pair. The present study has established the necessary genetic information to permit the unambiguous identification of 53 commonly marketed fish species in South Africa, the applications of which hold a plethora of benefits relating to ecology research, fisheries management and control of commercial practices.

Harris Andrew Steinman

Papers

A high incidence of species substitution and mislabelling detected in meat products sold in South Africa

DNA barcoding reveals a high incidence of fish species misrepresentation and substitution on the South African market

Evaluation of the 16S and 12S rRNA genes as universal markers for the identification of commercial fish species in South Africa.

Comparative study of different methods for the extraction of DNA from fish species commercially available in South Africa

Establishment of a mitochondrial DNA sequence database for the identification of fish species commercially available in South Africa