Perceptual computing is an application of Zadeh's theory of computing with words on the field of assisting people to make subjective judgments. == Perceptual computer == The perceptual computer – Per-C – an instantiation of perceptual computing – has the architecture that is depicted in Fig. 1 [2]–[6]. It consists of three components: encoder, CWW engine and decoder. Perceptions – words – activate the Per-C and are the Per-C output (along with data); so, it is possible for a human to interact with the Per-C using just a vocabulary. A vocabulary is application (context) dependent, and must be large enough so that it lets the end-user interact with the Per-C in a user-friendly manner. The encoder transforms words into fuzzy sets (FSs) and leads to a codebook – words with their associated FS models. The outputs of the encoder activate a Computing With Words (CWW) engine, whose output is one or more other FSs, which are then mapped by the decoder into a recommendation (subjective judgment) with supporting data. The recommendation may be in the form of a word, group of similar words, rank or class. Although many details are needed in order to implement the Per-C's three components – encoder, decoder and CWW engine – and they are covered in [5], it is when the Per-C is applied to specific applications, that the focus on the methodology becomes clear. Stepping back from those details, the methodology of perceptual computing is: Focus on an application (A). Establish a vocabulary (or vocabularies) for A. Collect interval end-point data from a group of subjects (representative of the subjects who will use the Per-C) for all of the words in the vocabulary. Map the collected word data into word-FOUs by using the Interval Approach [1], [5, Ch. 3]. The result of doing this is the codebook (or codebooks) for A, and completes the design of the encoder of the Per-C. Choose an appropriate CWW engine for A. It will map IT2 FSs into one or more IT2 FSs. Examples of CWW engines are: IF-THEN rules [5, Ch. 6] and Linguistic Weighted Averages [6], [5, Ch. 5]. If an existing CWW engine is available for A, then use its available mathematics to compute its output(s). Otherwise, develop such mathematics for the new kind of CWW engine. The new CWW engine should be constrained so that its output(s) resemble the FOUs in the codebook(s) for A. Map the IT2 FS outputs from the CWW engine into a recommendation at the output of the decoder. If the recommendation is a word, rank or class, then use existing mathematics to accomplish this mapping [5, Ch. 4]. Otherwise, develop such mathematics for the new kind of decoder. == Applications of Per-C == To-date a Per-C has been implemented for the following four applications: (1) investment decision-making, (2) social judgment making, (3) distributed decision making, and (4) hierarchical and distributed decision-making. A specific example of the fourth application is the so-called Journal Publication Judgment Advisor [5, Ch. 10] in which for the first time only words are used at every level of the following hierarchical and distributed decision making process: n reviewers have to provide a subjective recommendation about a journal article that has been sent to them by the Associate Editor, who then has to aggregate the independent recommendations into a final recommendation that is sent to the Editor-in-Chief of the journal. Because it is very problematic to ask reviewers to provide numerical scores for paper-evaluation sub-categories (the two major categories are Technical Merit and Presentation), such as importance, content, depth, style, organization, clarity, references, etc., each reviewer will only be asked to provide a linguistic score for each of these categories. They will not be asked for an overall recommendation about the paper because in the past it is quite common for reviewers who provide the same numerical scores for such categories to give very different publishing recommendations. By leaving a specific recommendation to the associate editor such inconsistencies can hope to be eliminated. How words can be aggregated to reflect each reviewer's recommendation as well as the expertise of each reviewer about the paper's subject matter is done using a linguistic weighted average. Although the journal publication judgment advisor uses reviewers and an associate editor, the word “reviewer” could be replaced by judge, expert, low-level manager, commander, referee, etc., and the term “associate editor” could be replaced by control center, command center, higher-level manager, etc. So, this application has potential wide applicability to many other applications. Recently, a new Per-C based Failure mode and effects analysis (FMEA) methodology was developed, with its application to edible bird's nest farming, in Borneo, has been reported. In addition, application of Per-C based method to educational assessment, for cooperative learning of students has been reported. In summary, the Per-C (whose development has taken more than a decade) is the first complete implementation of Zadeh's CWW paradigm, as applied to assisting people to make subjective judgments.
Multi-task learning
Multi-task learning (MTL) is a subfield of machine learning in which multiple learning tasks are solved at the same time, while exploiting commonalities and differences across tasks. This can result in improved learning efficiency and prediction accuracy for the task-specific models, when compared to training the models separately. Inherently, Multi-task learning is a multi-objective optimization problem having trade-offs between different tasks. Early versions of MTL were called "hints". In a widely cited 1997 paper, Rich Caruana gave the following characterization:Multitask Learning is an approach to inductive transfer that improves generalization by using the domain information contained in the training signals of related tasks as an inductive bias. It does this by learning tasks in parallel while using a shared representation; what is learned for each task can help other tasks be learned better. In the classification context, MTL aims to improve the performance of multiple classification tasks by learning them jointly. One example is a spam-filter, which can be treated as distinct but related classification tasks across different users. To make this more concrete, consider that different people have different distributions of features which distinguish spam emails from legitimate ones, for example an English speaker may find that all emails in Russian are spam, not so for Russian speakers. Yet there is a definite commonality in this classification task across users, for example one common feature might be text related to money transfer. Solving each user's spam classification problem jointly via MTL can let the solutions inform each other and improve performance. Further examples of settings for MTL include multiclass classification and multi-label classification. Multi-task learning works because regularization induced by requiring an algorithm to perform well on a related task can be superior to regularization that prevents overfitting by penalizing all complexity uniformly. One situation where MTL may be particularly helpful is if the tasks share significant commonalities and are generally slightly under sampled. However, as discussed below, MTL has also been shown to be beneficial for learning unrelated tasks. == Methods == The key challenge in multi-task learning, is how to combine learning signals from multiple tasks into a single model. This may strongly depend on how well different task agree with each other, or contradict each other. There are several ways to address this challenge: === Task grouping and overlap === Within the MTL paradigm, information can be shared across some or all of the tasks. Depending on the structure of task relatedness, one may want to share information selectively across the tasks. For example, tasks may be grouped or exist in a hierarchy, or be related according to some general metric. Suppose, as developed more formally below, that the parameter vector modeling each task is a linear combination of some underlying basis. Similarity in terms of this basis can indicate the relatedness of the tasks. For example, with sparsity, overlap of nonzero coefficients across tasks indicates commonality. A task grouping then corresponds to those tasks lying in a subspace generated by some subset of basis elements, where tasks in different groups may be disjoint or overlap arbitrarily in terms of their bases. Task relatedness can be imposed a priori or learned from the data. Hierarchical task relatedness can also be exploited implicitly without assuming a priori knowledge or learning relations explicitly. For example, the explicit learning of sample relevance across tasks can be done to guarantee the effectiveness of joint learning across multiple domains. === Exploiting unrelated tasks: Auxiliary learning === In auxiliary learning, one attempts learning a group of principal tasks using a group of auxiliary tasks, unrelated to the principal ones. With the right unrelated tasks, joint learning of unrelated tasks which use the same input data have been shown to be beneficial, and provide significant improvement over standard MTL. The reason is that prior knowledge about task relatedness can lead to sparser and more informative representations for each task grouping, essentially by screening out idiosyncrasies of the data distribution. It has been proposed to build on a prior multitask methodology by favoring a shared low-dimensional representation within each task grouping, and imposing a penalty on tasks from different groups which encourages the two representations to be orthogonal. Learning with auxiliary unrelated tasks poses two major challenges: Finding useful auxiliary tasks and combining losses of all tasks in a useful way. Some methods can learn these from data together with the training process, and combine tasks efficiently. === Transfer of knowledge === Related to multi-task learning is the concept of knowledge transfer. Whereas traditional multi-task learning implies that a shared representation is developed concurrently across tasks, transfer of knowledge implies a sequentially shared representation. Large scale machine learning projects such as the deep convolutional neural network GoogLeNet, an image-based object classifier, can develop robust representations which may be useful to further algorithms learning related tasks. For example, the pre-trained model can be used as a feature extractor to perform pre-processing for another learning algorithm. Or the pre-trained model can be used to initialize a model with similar architecture which is then fine-tuned to learn a different classification task. === Multiple non-stationary tasks === Traditionally Multi-task learning and transfer of knowledge are applied to stationary learning settings. Their extension to non-stationary environments is termed Group online adaptive learning (GOAL). Sharing information could be particularly useful if learners operate in continuously changing environments, because a learner could benefit from previous experience of another learner to quickly adapt to their new environment. Such group-adaptive learning has numerous applications, from predicting financial time-series, through content recommendation systems, to visual understanding for adaptive autonomous agents. === Multi-task optimization === Multi-task optimization focuses on solving optimizing the whole process. The paradigm has been inspired by the well-established concepts of transfer learning and multi-task learning in predictive analytics. The key motivation behind multi-task optimization is that if optimization tasks are related to each other in terms of their optimal solutions or the general characteristics of their function landscapes, the search progress can be transferred to substantially accelerate the search on the other. The success of the paradigm is not necessarily limited to one-way knowledge transfers from simpler to more complex tasks. In practice an attempt is to intentionally solve a more difficult task that may unintentionally solve several smaller problems. There is a direct relationship between multitask optimization and multi-objective optimization. In some cases, the simultaneous training of seemingly related tasks may hinder performance compared to single-task models. Commonly, MTL models employ task-specific modules on top of a joint feature representation obtained using a shared module. Since this joint representation must capture useful features across all tasks, MTL may hinder individual task performance if the different tasks seek conflicting representation, i.e., the gradients of different tasks point to opposing directions or differ significantly in magnitude. This phenomenon is commonly referred to as negative transfer. To mitigate this issue, various MTL optimization methods have been proposed. It has been reported that meta-knowledge transfer could help avoid negative transfer.Besides, the per-task gradients are combined into a joint update direction through various aggregation algorithms or heuristics. There are several common approaches for multi-task optimization: Bayesian optimization, evolutionary computation, and approaches based on Game theory. ==== Multi-task Bayesian optimization ==== Multi-task Bayesian optimization is a modern model-based approach that leverages the concept of knowledge transfer to speed up the automatic hyperparameter optimization process of machine learning algorithms. The method builds a multi-task Gaussian process model on the data originating from different searches progressing in tandem. The captured inter-task dependencies are thereafter utilized to better inform the subsequent sampling of candidate solutions in respective search spaces. ==== Evolutionary multi-tasking ==== Evolutionary multi-tasking has been explored as a means of exploiting the implicit parallelism of population-based search algorithms to simultaneously progress multiple distinct optimization tasks. By mapping all task
Ontology-based data integration
Ontology-based data integration involves the use of one or more ontologies to effectively combine data or information from multiple heterogeneous sources. It is one of the multiple data integration approaches and may be classified as Global-As-View (GAV). The effectiveness of ontology‑based data integration is closely tied to the consistency and expressivity of the ontology used in the integration process. == Background == Data from multiple sources are characterized by multiple types of heterogeneity. The following hierarchy is often used: Syntactic heterogeneity: is a result of differences in representation format of data Schematic or structural heterogeneity: the native model or structure to store data differ in data sources leading to structural heterogeneity. Schematic heterogeneity that particularly appears in structured databases is also an aspect of structural heterogeneity. Semantic heterogeneity: differences in interpretation of the 'meaning' of data are source of semantic heterogeneity System heterogeneity: use of different operating system, hardware platforms lead to system heterogeneity Ontologies, as formal models of representation with explicitly defined concepts and named relationships linking them, are used to address the issue of semantic heterogeneity in data sources. In domains like bioinformatics and biomedicine, the rapid development, adoption and public availability of ontologies [1] Archived 2007-06-16 at the Wayback Machine has made it possible for the data integration community to leverage them for semantic integration of data and information. == The role of ontologies == Ontologies enable the unambiguous identification of entities in heterogeneous information systems and assertion of applicable named relationships that connect these entities together. Specifically, ontologies play the following roles: Content Explication The ontology enables accurate interpretation of data from multiple sources through the explicit definition of terms and relationships in the ontology. Query Model In some systems like SIMS, the query is formulated using the ontology as a global query schema. Verification The ontology verifies the mappings used to integrate data from multiple sources. These mappings may either be user specified or generated by a system. == Approaches using ontologies for data integration == There are three main architectures that are implemented in ontology‑based data integration applications, namely, Single ontology approach A single ontology is used as a global reference model in the system. This is the simplest approach as it can be simulated by other approaches. SIMS is a prominent example of this approach. The Structured Knowledge Source Integration component of Research Cyc is another prominent example of this approach. (Title = Harnessing Cyc to Answer Clinical Researchers' Ad Hoc Queries). The Gellish Taxonomic Dictionary-Ontology follows this approach as well. Multiple ontologies Multiple ontologies, each modeling an individual data source, are used in combination for integration. Though, this approach is more flexible than the single ontology approach, it requires creation of mappings between the multiple ontologies. Ontology mapping is a challenging issue and is focus of large number of research efforts in computer science [2]. The OBSERVER system is an example of this approach. Hybrid approaches The hybrid approach involves the use of multiple ontologies that subscribe to a common, top-level vocabulary. The top-level vocabulary defines the basic terms of the domain. Thus, the hybrid approach makes it easier to use multiple ontologies for integration in presence of the common vocabulary.
Token-based replay
Token-based replay technique is a conformance checking algorithm that checks how well a process conforms with its model by replaying each trace on the model (in Petri net notation ). Using the four counters produced tokens, consumed tokens, missing tokens, and remaining tokens, it records the situations where a transition is forced to fire and the remaining tokens after the replay ends. Based on the count at each counter, we can compute the fitness value between the trace and the model. == The algorithm == Source: The token-replay technique uses four counters to keep track of a trace during the replaying: p: Produced tokens c: Consumed tokens m: Missing tokens (consumed while not there) r: Remaining tokens (produced but not consumed) Invariants: At any time: p + m ≥ c ≥ m {\displaystyle p+m\geq c\geq m} At the end: r = p + m − c {\displaystyle r=p+m-c} At the beginning, a token is produced for the source place (p = 1) and at the end, a token is consumed from the sink place (c' = c + 1). When the replay ends, the fitness value can be computed as follows: 1 2 ( 1 − m c ) + 1 2 ( 1 − r p ) {\displaystyle {\frac {1}{2}}(1-{\frac {m}{c}})+{\frac {1}{2}}(1-{\frac {r}{p}})} == Example == Suppose there is a process model in Petri net notation as follows: === Example 1: Replay the trace (a, b, c, d) on the model M === Step 1: A token is initiated. There is one produced token ( p = 1 {\displaystyle p=1} ). Step 2: The activity a {\displaystyle \mathbf {a} } consumes 1 token to be fired and produces 2 tokens ( p = 1 + 2 = 3 {\displaystyle p=1+2=3} and c = 1 {\displaystyle c=1} ). Step 3: The activity b {\displaystyle \mathbf {b} } consumes 1 token and produces 1 token ( p = 3 + 1 = 4 {\displaystyle p=3+1=4} and c = 1 + 1 = 2 {\displaystyle c=1+1=2} ). Step 4: The activity c {\displaystyle \mathbf {c} } consumes 1 token and produces 1 token ( p = 4 + 1 = 5 {\displaystyle p=4+1=5} and c = 2 + 1 = 3 {\displaystyle c=2+1=3} ). Step 5: The activity d {\displaystyle \mathbf {d} } consumes 2 tokens and produces 1 token ( p = 5 + 1 = 6 {\displaystyle p=5+1=6} , c = 3 + 2 = 5 {\displaystyle c=3+2=5} ). Step 6: The token at the end place is consumed ( c = 5 + 1 = 6 {\displaystyle c=5+1=6} ). The trace is complete. The fitness of the trace ( a , b , c , d {\displaystyle \mathbf {a,b,c,d} } ) on the model M {\displaystyle \mathbf {M} } is: 1 2 ( 1 − m c ) + 1 2 ( 1 − r p ) = 1 2 ( 1 − 0 6 ) + 1 2 ( 1 − 0 6 ) = 1 {\displaystyle {\frac {1}{2}}(1-{\frac {m}{c}})+{\frac {1}{2}}(1-{\frac {r}{p}})={\frac {1}{2}}(1-{\frac {0}{6}})+{\frac {1}{2}}(1-{\frac {0}{6}})=1} === Example 2: Replay the trace (a, b, d) on the model M === Step 1: A token is initiated. There is one produced token ( p = 1 {\displaystyle p=1} ). Step 2: The activity a {\displaystyle \mathbf {a} } consumes 1 token to be fired and produces 2 tokens ( p = 1 + 2 = 3 {\displaystyle p=1+2=3} and c = 1 {\displaystyle c=1} ). Step 3: The activity b {\displaystyle \mathbf {b} } consumes 1 token and produces 1 token ( p = 3 + 1 = 4 {\displaystyle p=3+1=4} and c = 1 + 1 = 2 {\displaystyle c=1+1=2} ). Step 4: The activity d {\displaystyle \mathbf {d} } needs to be fired but there are not enough tokens. One artificial token was produced and the missing token counter is increased by one ( m = 1 {\displaystyle m=1} ). The artificial token and the token at place [ b , d ] {\displaystyle [\mathbf {b,d} ]} are consumed ( c = 2 + 2 = 4 {\displaystyle c=2+2=4} ) and one token is produced at place end ( p = 4 + 1 = 5 {\displaystyle p=4+1=5} ). Step 5: The token in the end place is consumed ( c = 4 + 1 = 5 {\displaystyle c=4+1=5} ). The trace is complete. There is one remaining token at place [ a , c ] {\displaystyle [\mathbf {a,c} ]} ( r = 1 {\displaystyle r=1} ). The fitness of the trace ( a , b , d {\displaystyle \mathbf {a,b,d} } ) on the model M {\displaystyle \mathbf {M} } is: 1 2 ( 1 − m c ) + 1 2 ( 1 − r p ) = 1 2 ( 1 − 1 5 ) + 1 2 ( 1 − 1 5 ) = 0.8 {\displaystyle {\frac {1}{2}}(1-{\frac {m}{c}})+{\frac {1}{2}}(1-{\frac {r}{p}})={\frac {1}{2}}(1-{\frac {1}{5}})+{\frac {1}{2}}(1-{\frac {1}{5}})=0.8}
Flajolet–Martin algorithm
The Flajolet–Martin algorithm is an algorithm for approximating the number of distinct elements in a stream with a single pass and space-consumption logarithmic in the maximal number of possible distinct elements in the stream (the count-distinct problem). The algorithm was introduced by Philippe Flajolet and G. Nigel Martin in their 1984 article "Probabilistic Counting Algorithms for Data Base Applications". Later it has been refined in "LogLog counting of large cardinalities" by Marianne Durand and Philippe Flajolet, and "HyperLogLog: The analysis of a near-optimal cardinality estimation algorithm" by Philippe Flajolet et al. In their 2010 article "An optimal algorithm for the distinct elements problem", Daniel M. Kane, Jelani Nelson and David P. Woodruff give an improved algorithm, which uses nearly optimal space and has optimal O(1) update and reporting times. == The algorithm == Assume that we are given a hash function h a s h ( x ) {\displaystyle \mathrm {hash} (x)} that maps input x {\displaystyle x} to integers in the range [ 0 ; 2 L − 1 ] {\displaystyle [0;2^{L}-1]} , and where the outputs are sufficiently uniformly distributed. Note that the set of integers from 0 to 2 L − 1 {\displaystyle 2^{L}-1} corresponds to the set of binary strings of length L {\displaystyle L} . For any non-negative integer y {\displaystyle y} , define b i t ( y , k ) {\displaystyle \mathrm {bit} (y,k)} to be the k {\displaystyle k} -th bit in the binary representation of y {\displaystyle y} , such that: y = ∑ k ≥ 0 b i t ( y , k ) 2 k . {\displaystyle y=\sum _{k\geq 0}\mathrm {bit} (y,k)2^{k}.} We then define a function ρ ( y ) {\displaystyle \rho (y)} that outputs the position of the least-significant set bit in the binary representation of y {\displaystyle y} , and L {\displaystyle L} if no such set bit can be found as all bits are zero: ρ ( y ) = { min { k ≥ 0 ∣ b i t ( y , k ) ≠ 0 } y > 0 L y = 0 {\displaystyle \rho (y)={\begin{cases}\min\{k\geq 0\mid \mathrm {bit} (y,k)\neq 0\}&y>0\\L&y=0\end{cases}}} Note that with the above definition we are using 0-indexing for the positions, starting from the least significant bit. For example, ρ ( 13 ) = ρ ( 1101 2 ) = 0 {\displaystyle \rho (13)=\rho (1101_{2})=0} , since the least significant bit is a 1 (0th position), and ρ ( 8 ) = ρ ( 1000 2 ) = 3 {\displaystyle \rho (8)=\rho (1000_{2})=3} , since the least significant set bit is at the 3rd position. At this point, note that under the assumption that the output of our hash function is uniformly distributed, then the probability of observing a hash output ending with 2 k {\displaystyle 2^{k}} (a one, followed by k {\displaystyle k} zeroes) is 2 − ( k + 1 ) {\displaystyle 2^{-(k+1)}} , since this corresponds to flipping k {\displaystyle k} heads and then a tail with a fair coin. Now the Flajolet–Martin algorithm for estimating the cardinality of a multiset M {\displaystyle M} is as follows: Initialize a bit-vector BITMAP to be of length L {\displaystyle L} and contain all 0s. For each element x {\displaystyle x} in M {\displaystyle M} : Calculate the index i = ρ ( h a s h ( x ) ) {\displaystyle i=\rho (\mathrm {hash} (x))} . Set B I T M A P [ i ] = 1 {\displaystyle \mathrm {BITMAP} [i]=1} . Let R {\displaystyle R} denote the smallest index i {\displaystyle i} such that B I T M A P [ i ] = 0 {\displaystyle \mathrm {BITMAP} [i]=0} . Estimate the cardinality of M {\displaystyle M} as 2 R / ϕ {\displaystyle 2^{R}/\phi } , where ϕ ≈ 0.77351 {\displaystyle \phi \approx 0.77351} . The idea is that if n {\displaystyle n} is the number of distinct elements in the multiset M {\displaystyle M} , then B I T M A P [ 0 ] {\displaystyle \mathrm {BITMAP} [0]} is accessed approximately n / 2 {\displaystyle n/2} times, B I T M A P [ 1 ] {\displaystyle \mathrm {BITMAP} [1]} is accessed approximately n / 4 {\displaystyle n/4} times and so on. Consequently, if i ≫ log 2 n {\displaystyle i\gg \log _{2}n} , then B I T M A P [ i ] {\displaystyle \mathrm {BITMAP} [i]} is almost certainly 0, and if i ≪ log 2 n {\displaystyle i\ll \log _{2}n} , then B I T M A P [ i ] {\displaystyle \mathrm {BITMAP} [i]} is almost certainly 1. If i ≈ log 2 n {\displaystyle i\approx \log _{2}n} , then B I T M A P [ i ] {\displaystyle \mathrm {BITMAP} [i]} can be expected to be either 1 or 0. The correction factor ϕ ≈ 0.77351 {\displaystyle \phi \approx 0.77351} (OEIS: A244256) is found by calculations, which can be found in the original article. == Improving accuracy == A problem with the Flajolet–Martin algorithm in the above form is that the results vary significantly. A common solution has been to run the algorithm multiple times with k {\displaystyle k} different hash functions and combine the results from the different runs. One idea is to take the mean of the k {\displaystyle k} results together from each hash function, obtaining a single estimate of the cardinality. The problem with this is that averaging is very susceptible to outliers (which are likely here). A different idea is to use the median, which is less prone to be influences by outliers. The problem with this is that the results can only take form 2 R / ϕ {\displaystyle 2^{R}/\phi } , where R {\displaystyle R} is integer. A common solution is to combine both the mean and the median: Create k ⋅ l {\displaystyle k\cdot l} hash functions and split them into k {\displaystyle k} distinct groups (each of size l {\displaystyle l} ). Within each group use the mean for aggregating together the l {\displaystyle l} results, and finally take the median of the k {\displaystyle k} group estimates as the final estimate. The 2007 HyperLogLog algorithm splits the multiset into subsets and estimates their cardinalities, then it uses the harmonic mean to combine them into an estimate for the original cardinality.
DoorDash
DoorDash, Inc. is an American company operating online food ordering and food delivery. It trades under the symbol DASH. With a 56% market share, DoorDash is the largest food delivery platform in the United States. It also has a 60% market share in the convenience delivery category. As of December 31, 2020, the platform was used by 450,000 merchants, 20 million consumers, and had over one million delivery couriers. Founded by Tony Xu, Andy Fang, Stanley Tang and Evan Moore, DoorDash made its debut on the Fortune 500 list in 2024, ranking No. 443. DoorDash has been sued for or held legally liable for withholding tips, reducing tip transparency, antitrust price manipulation, listing restaurants without permission, misclassifying workers, withholding sick time, and illegally selling personal data. As of April 2026, DoorDash operates in the United States (including Puerto Rico), Canada, Australia, and New Zealand. Through its subsidiaries Deliveroo and Wolt, the company also operates across Europe, as well as in Azerbaijan, Georgia, Israel, Kazakhstan, Kuwait, and the United Arab Emirates. == History == In January 2013, Stanford University students Tony Xu, Stanley Tang, Andy Fang and Evan Moore launched PaloAltoDelivery.com in Palo Alto, California. In the summer of 2013, it received US$120,000 in seed money from Y Combinator in exchange for a 7% stake. It incorporated as DoorDash in June 2013. DoorDash's first partnership with a fast food burger restaurant chain was in April 2016, when it partnered with CKE Restaurants, parent company of Carl's Jr. and Hardee's, for food delivery. In December 2017, DoorDash announced its partnership with Wendy's for delivery from its restaurants. In December 2018, DoorDash overtook Uber Eats to hold the second position in total US food delivery sales, behind GrubHub. By March 2019, it had exceeded GrubHub in total sales, at 27.6% of the on-demand delivery market. By early 2019, DoorDash was the largest food delivery provider in the U.S., as measured by consumer spending. In October 2019, DoorDash opened its first ghost kitchen, DoorDash Kitchen, in Redwood City, California, with four restaurants operating at the location. By June 2020, DoorDash had raised more than $2.5 billion over several financing rounds from investors including Y Combinator, Charles River Ventures, SV Angel, Khosla Ventures, Sequoia Capital, SoftBank Group, GIC, and Kleiner Perkins. DoorDash announced a partnership with KFC in September 2020, followed by Taco Bell in October 2020. In November 2020, DoorDash announced the opening of its first physical restaurant location, partnering up with Bay Area restaurant Burma Bites to offer delivery and pick-up orders. In December 2020, it became a public company via an initial public offering, raising $3.37 billion. In November 2021, DoorDash acquired Finland's Wolt for €7bn. In August 2022, DoorDash announced it would end its partnership with Walmart in September, ending the companies' cooperation agreement from 2018. In November 2022, DoorDash announced plans to lay off 1,250 corporate employees, or about six percent of its workforce, to rein in expenses. In June 2023, DoorDash announced it would give its drivers the option of earning an hourly minimum wage instead of being paid per delivery. However, drivers are only paid hourly when on an active delivery. In September 2023, the company transferred its stock listing from the New York Stock Exchange to the Nasdaq. On December 18, 2023, DoorDash was added to the Nasdaq-100 index. In March 2025, DoorDash announced a partnership with Klarna, a Buy Now, Pay Later (BNPL) service, letting customers schedule small payments over a set period of time. DoorDash received widespread criticism from this decision, including internet mockery, given concerns about the increase of household debt in America. In 2025, DoorDash acquired the UK-based delivery service Deliveroo for $3.88 billion. The combined company operates in 40 countries and serves 50 million users monthly. In September 2025, DoorDash and Ace Hardware (the largest hardware cooperative) announced their partnership to offer delivery for home use products from over 4,000 Ace locations. == Lawsuits against DoorDash == === 2017 class-action lawsuit for misclassifying workers === In 2017, a class-action lawsuit was filed against DoorDash for allegedly misclassifying delivery drivers in California and Massachusetts as independent contractors. In 2022, a tentative settlement was reached in which DoorDash would pay $100 million total, with $61 million going to over 900,000 drivers, paying out just over $130 per driver, and $28 million for the lawyers. Gizmodo criticized the settlement, noting that the $413 million that DoorDash CEO Tony Xu received the previous year was one of the largest CEO compensation packages of all time. === 2019 data breach lawsuit === On May 4, 2019, DoorDash confirmed 4.9 million customers, delivery workers and merchants had sensitive information stolen via a data breach. Those who joined the platform after April 5, 2018, were unaffected by the breach. A class-action lawsuit for the breach was filed against DoorDash in October 2019. === Withholding of tips and subsequent class-action lawsuits === In July 2019, the company's tipping policy was criticized by The New York Times, and later The Verge and Vox and Gothamist. Drivers receive a guaranteed minimum per order that is paid by DoorDash by default. When a customer added a tip, instead of going directly to the driver, it first went to the company to cover the guaranteed minimum. Drivers then only directly received the part of the tip that exceeded the guaranteed minimum per order. In January 2020, it was reported that DoorDash had lied about skimming tips from its drivers, causing them to earn an average of $1.45 an hour after expenses, and that after the company had allegedly overhauled its tipping system, DoorDash was still manipulating per-delivery payouts at the expense of drivers. A DoorDash customer filed a class action lawsuit against the company for its "materially false and misleading" tipping policy. The case was referred to arbitration in August 2020. Under pressure, the company revised its policy. The company settled a lawsuit with District of Columbia Attorney General Karl Racine for $2.5 million, with funds going to deliverers, the government, and to charity. ==== 2021 driver strike for tip transparency ==== In July 2021, DoorDash drivers went on strike to protest lack of tip transparency and to ask for higher pay. At the time of the strike, and, as of June 2022, DoorDash did not allow drivers to see the full tip amounts prior to accepting a delivery in the app. If customers tip over a set amount for the order total, Doordash hides a portion of the tip until the delivery is complete. The strike occurred after DoorDash rewrote its code to cut off access to Para, a third-party app that drivers had been using to see the full tip amounts. ==== 2025 class-action lawsuit settlement ==== In 2025, DoorDash agreed to pay around $17 million for "misleading both consumers and delivery workers" with tips being docked from drivers' pay instead of directly going to drivers. === 2020 antitrust litigation === In April 2020, in the case of Davitashvili v. GrubHub Inc. DoorDash, Grubhub, Postmates, and Uber Eats were accused of monopolistic power by only listing restaurants on its apps if the restaurant owners signed contracts which include clauses that require prices be the same for dine-in customers as for customers receiving delivery. The plaintiffs stated that this arrangement increases the cost for dine-in customers, as they are required to subsidize the cost of delivery; and that the apps charge "exorbitant" fees, which range from 13% to 40% of revenue, while the average restaurant's profit ranges from 3% to 9% of revenue. The lawsuit seeks treble damages, including for overcharges, since April 14, 2016, for dine-in and delivery customers in the United States at restaurants using the defendants’ delivery apps. Although several preliminary documents in the case have now been filed, a trial date has not yet been set. === Litigation for illegal unauthorized restaurant listing === In May 2021, DoorDash was criticized for unauthorized listings of restaurants who had not given permission to appear on the app. The company was sued by Lona's Lil Eats in St. Louis, with the lawsuit claiming that DoorDash had listed them without permission, then prevented any orders to the restaurant from going through and redirecting customers to other restaurants instead, because Lona's was "too far away," when in reality it had not paid DoorDash a fee for listing. This aspect of DoorDash's business practice is illegal in California. === 2021 lawsuit by the city of Chicago === In August 2021, the city of Chicago sued DoorDash and GrubHub. According to Chicago mayor Lori Lightfoot, the companies broke the law by using "unfair and deceptive t
Record linkage
Record linkage (also known as data matching, data linkage, entity resolution, and many other terms) is the task of finding records in a data set that refer to the same entity across different data sources (e.g., data files, books, websites, and databases). Record linkage is necessary when joining different data sets based on entities that may or may not share a common identifier (e.g., database key, URI, National identification number), which may be due to differences in record shape, storage location, or curator style or preference. A data set that has undergone RL-oriented reconciliation may be referred to as being cross-linked. == Naming conventions == "Record linkage" is the term used by statisticians, epidemiologists, and historians, among others, to describe the process of joining records from one data source with another that describe the same entity. However, many other terms are used for this process. Unfortunately, this profusion of terminology has led to few cross-references between these research communities. Computer scientists often refer to it as "data matching" or as the "object identity problem". Commercial mail and database applications refer to it as "merge/purge processing" or "list washing". Other names used to describe the same concept include: "coreference/entity/identity/name/record resolution", "entity disambiguation/linking", "fuzzy matching", "duplicate detection", "deduplication", "record matching", "(reference) reconciliation", "object identification", "data/information integration" and "conflation". While they share similar names, record linkage and linked data are two separate approaches to processing and structuring data. Although both involve identifying matching entities across different data sets, record linkage standardly equates "entities" with human individuals; by contrast, Linked Data is based on the possibility of interlinking any web resource across data sets, using a correspondingly broader concept of identifier, namely a URI. == History == The initial idea of record linkage goes back to Halbert L. Dunn in his 1946 article titled "Record Linkage" published in the American Journal of Public Health. Howard Borden Newcombe then laid the probabilistic foundations of modern record linkage theory in a 1959 article in Science. These were formalized in 1969 by Ivan Fellegi and Alan Sunter, in their pioneering work "A Theory For Record Linkage", where they proved that the probabilistic decision rule they described was optimal when the comparison attributes were conditionally independent. In their work they recognized the growing interest in applying advances in computing and automation to large collections of administrative data, and the Fellegi-Sunter theory remains the mathematical foundation for many record linkage applications. Since the late 1990s, various machine learning techniques have been developed that can, under favorable conditions, be used to estimate the conditional probabilities required by the Fellegi-Sunter theory. Several researchers have reported that the conditional independence assumption of the Fellegi-Sunter algorithm is often violated in practice; however, published efforts to explicitly model the conditional dependencies among the comparison attributes have not resulted in an improvement in record linkage quality. On the other hand, machine learning or neural network algorithms that do not rely on these assumptions often provide far higher accuracy, when sufficient labeled training data is available. Record linkage can be done entirely without the aid of a computer, but the primary reasons computers are often used to complete record linkages are to reduce or eliminate manual review and to make results more easily reproducible. Computer matching has the advantages of allowing central supervision of processing, better quality control, speed, consistency, and better reproducibility of results. == Methods == === Data preprocessing === Record linkage is highly sensitive to the quality of the data being linked, so all data sets under consideration (particularly their key identifier fields) should ideally undergo a data quality assessment before record linkage. Many key identifiers for the same entity can be presented quite differently between (and even within) data sets, which can greatly complicate record linkage unless understood ahead of time. For example, key identifiers for a man named William J. Smith might appear in three different data sets as follows: In this example, the different formatting styles lead to records that look different but in fact all refer to the same entity with the same logical identifier values. Most, if not all, record linkage strategies would result in more accurate linkage if these values were first normalized or standardized into a consistent format (e.g., all names are "Surname, Given name", and all dates are "YYYY/MM/DD"). Standardization can be accomplished through simple rule-based data transformations or more complex procedures such as lexicon-based tokenization and probabilistic hidden Markov models. Several of the packages listed in the Software Implementations section provide some of these features to simplify the process of data standardization. === Entity resolution === Entity resolution is an operational intelligence process, typically powered by an entity resolution engine or middleware, whereby organizations can connect disparate data sources with a view to understand possible entity matches and non-obvious relationships across multiple data silos. It analyzes all of the information relating to individuals and/or entities from multiple sources of data, and then applies likelihood and probability scoring to determine which identities are a match and what, if any, non-obvious relationships exist between those identities. Entity resolution engines are typically used to uncover risk, fraud, and conflicts of interest, but are also useful tools for use within customer data integration (CDI) and master data management (MDM) requirements. Typical uses for entity resolution engines include terrorist screening, insurance fraud detection, USA Patriot Act compliance, organized retail crime ring detection and applicant screening. For example, across different data silos – employee records, vendor data, watch lists, etc. – an organization may have several variations of an entity named ABC, which may or may not be the same individual. These entries may, in fact, appear as ABC1, ABC2, or ABC3 within those data sources. By comparing similarities between underlying attributes such as address, date of birth, or social security number, the user can eliminate some possible matches and confirm others as very likely matches. Entity resolution engines then apply rules, based on common sense logic, to identify hidden relationships across the data. In the example above, perhaps ABC1 and ABC2 are not the same individual, but rather two distinct people who share common attributes such as address or phone number. ==== Data matching ==== While entity resolution solutions include data matching technology, many data matching offerings do not fit the definition of entity resolution. Here are four factors that distinguish entity resolution from data matching, according to John Talburt, director of the UALR Center for Advanced Research in Entity Resolution and Information Quality: Works with both structured and unstructured records, and it entails the process of extracting references when the sources are unstructured or semi-structured Uses elaborate business rules and concept models to deal with missing, conflicting, and corrupted information Utilizes non-matching, asserted linking (associate) information in addition to direct matching Uncovers non-obvious relationships and association networks (i.e. who's associated with whom) In contrast to data quality products, more powerful identity resolution engines also include a rules engine and workflow process, which apply business intelligence to the resolved identities and their relationships. These advanced technologies make automated decisions and impact business processes in real time, limiting the need for human intervention. === Deterministic record linkage === The simplest kind of record linkage, called deterministic or rules-based record linkage, generates links based on the number of individual identifiers that match among the available data sets. Two records are said to match via a deterministic record linkage procedure if all or some identifiers (above a certain threshold) are identical. Deterministic record linkage is a good option when the entities in the data sets are identified by a common identifier, or when there are several representative identifiers (e.g., name, date of birth, and sex when identifying a person) whose quality of data is relatively high. As an example, consider two standardized data sets, Set A and Set B, that contain different bits of information about patients in a hospital system. T