Abstract: A method, computer system, and a computer program product for semi-automated exam grading are provided. An exam that includes a question is received. A model answer to the question is received. The model answer includes an essay form. A computer system identifies constructs in the model answer. An examinee answer sheet that includes an examinee answer to the question is received. The computer system grades the examinee answer of the examinee answer sheet based on the constructs of the model answer. The computer system presents the graded examinee answer sheet to an evaluator for review. The graded examinee answer sheet includes one or more distinction markers along with the examinee answer. The one or more distinction markers indicate respective priority levels of corresponding portions of the examinee answer.

Abstract: A method and system of constructing a network map from imagery comprising using an iterative search process guided by a CNN-based decision function to derive a network graph directly from the output of the CNN.

Abstract: The present disclosure provides a cargo revenue management system and method that increases the efficiency of cargo revenue management by increasing the prediction accuracy of cargo volumes that customers will tender in order to generate more efficient decisions to accept or reject cargo bookings. The provided system accomplishes this increased efficiency by identifying cargo volumes that customers arbitrarily book when an actual volume is unknown as disguised missing values and deemphasizing such values in the prediction of a cargo volume that will be received. The provided system additionally utilizes machine-learning models trained on a combination of features to predict a cargo volume that will be received for a particular cargo booking. Based on the predicted cargo volume that will be received, the system generates a decision of whether to accept or reject the cargo booking to maximize revenue generation.

Abstract: A global effort is underway to identify compounds to treat emerging virus infections, such as COVID-19. Since de novo compound design is an extremely long, time-consuming, and expensive process, efforts are underway to discover existing compounds that can be repurposed for COVID-19 and new viral diseases. The present invention discloses a machine learning representation framework that uses deep learning-induced vector embeddings of compounds and viral proteins as features to predict compound-viral protein activity. The prediction model uses a consensus framework to rank approved compounds against viral proteins of interest.

Abstract: A method, computer system, and a computer program product for semi-automated exam grading are provided. An exam that includes a question is received. A model answer to the question is received. The model answer includes an essay form. A computer system identifies constructs in the model answer. An examinee answer sheet that includes an examinee answer to the question is received. The computer system grades the examinee answer of the examinee answer sheet based on the constructs of the model answer. The computer system presents the graded examinee answer sheet to an evaluator for review. The graded examinee answer sheet includes one or more distinction markers along with the examinee answer. The one or more distinction markers indicate respective priority levels of corresponding portions of the examinee answer.

Abstract: A method for identifying aircraft faults, comprising: receiving a dataset comprising a plurality of low priority messages and a plurality of high priority messages, each low priority message identifying a minor aircraft fault and each high priority message identifying a major aircraft fault; for each low priority message, generating an embedding vector which maps the low priority message in an embedding space; for each high priority message, generating an embedding vector which maps the high priority message in the embedding space; providing, to a machine learning unit, the embedding vector for each low priority message of the plurality of low priority messages and the embedding vector for each high priority message of the plurality of high priority messages; and obtaining, from the machine learning unit, a probability of a target high priority message occurring based on each low priority message of the plurality of low priority messages.

Abstract: A heater core assembly (10) comprising: a core (12) comprising a plurality of micro-tubes (13A, 13B), the plurality of micro-tubes (13A, 13B) being stacked in horizontal rows (15) between at least two headers (18) by inserting ends of each of the micro-tubes (13A,13B) into slots (42A, 42B) provided in the headers (18); a partition plate (30) disposed vertically in each of header (18) to define two vertical chambers (18A, 18B); wherein each of the horizontal rows (15) include at least one first micro-tube (13A) inserted in the first chamber (18A) and at least second micro-tube (13B) inserted in the second chamber (18B) to enable flow of the coolant in the core assembly (10).

Abstract: A heat exchanger (100) having a plurality of plates (101) manufactured preferably but not limiting to the stamping process, the plates has been configured to accommodate the internal fins (106). The plates (101) also define plurality of the passages (102) for flowing at least two fluids. A plurality of conduits (103) fluidly coupled to a first end and second of the end of the plates (101) which allows the flow of the fluids. At least one inlet (104) coupled to a first end, and at least one outlet (105) coupled to the second end of the plurality of plates (101) configured to allow the flow of the fluids wherein each fluid flow in a different direction from the other, a plurality of inner fins (106) disposed on a surface of each of the plurality of plates (101) for increasing the surface to volume ratio of the first and second fluid to achieve pre-defined thermal performance.

Abstract: The present disclosure generally relates to a cost-based optimizer for efficiently processing data through the use of multiple different data processing platforms. The cost-based optimizer may receive an input plan for processing data that includes a number of base operators. The cost-based optimizer may then determine execution operators for each base operator, where each execution operator corresponds to a different data processing platform. From the execution operators, the cost-based optimizer may determine possible subplans for executing the input plan on one or more data processing platforms. The cost-based optimizer may determine the cost of executing each possible subplan and choose the subplan with the lowest cost.

Abstract: A method for identifying aircraft faults, comprising: receiving a dataset comprising a plurality of low priority messages and a plurality of high priority messages, each low priority message identifying a minor aircraft fault and each high priority message identifying a major aircraft fault; for each low priority message, generating an embedding vector which maps the low priority message in an embedding space; for each high priority message, generating an embedding vector which maps the high priority message in the embedding space; providing, to a machine learning unit, the embedding vector for each low priority message of the plurality of low priority messages and the embedding vector for each high priority message of the plurality of high priority messages; and obtaining, from the machine learning unit, a probability of a target high priority message occurring based on each low priority message of the plurality of low priority messages.

Abstract: A method for identifying aircraft faults, comprising: receiving aircraft health dataset comprising plurality of maintenance identifiers which each identify aircraft fault; storing diagnostics database storing plurality of part identifiers which each identify part of aircraft which is possible cause of generation of at least one maintenance identifier; generating graph of plurality of maintenance identifiers and plurality of edges in which maintenance identifiers are connected to one another by edge if maintenance identifiers are identified by common part identifier in diagnostics database; extracting clique from graph, clique comprising plurality of maintenance identifiers and respective plurality of edges of graph; determining intersection between at least two edges of clique; identifying candidate part identifier which is common to intersecting edges of clique, candidate part identifier identifying part of aircraft which is possible cause of generation of at least some of maintenance identifiers of clique;

Abstract: The present disclosure provides a cargo revenue management system and method that increases the efficiency of cargo revenue management by increasing the prediction accuracy of cargo volumes that customers will tender in order to generate more efficient decisions to accept or reject cargo bookings. The provided system accomplishes this increased efficiency by identifying cargo volumes that customers arbitrarily book when an actual volume is unknown as disguised missing values and deemphasizing such values in the prediction of a cargo volume that will be received. The provided system additionally utilizes machine-learning models trained on a combination of features to predict a cargo volume that will be received for a particular cargo booking. Based on the predicted cargo volume that will be received, the system generates a decision of whether to accept or reject the cargo booking to maximize revenue generation.

Abstract: A method and system of constructing a network map from imagery comprising using an iterative search process guided by a CNN-based decision function to derive a network graph directly from the output of the CNN.

Abstract: The present disclosure generally relates to a cost-based optimizer for efficiently processing data through the use of multiple different data processing platforms. The cost-based optimizer may receive an input plan for processing data that includes a number of base operators. The cost-based optimizer may then determine execution operators for each base operator, where each execution operator corresponds to a different data processing platform. From the execution operators, the cost-based optimizer may determine possible subplans for executing the input plan on one or more data processing platforms. The cost-based optimizer may determine the cost of executing each possible subplan and choose the subplan with the lowest cost.